> > 🚨 Qwen2.5-Math mainly supports solving English and Chinese math problems through CoT and TIR. We do not recommend using this series of models for other tasks. > >
","release_date":"2024-09-19","context_window":4096,"license":"apache-2.0","tagline":"> [!Warning] > > > 🚨 Qwen2.5-Math mainly supports solving English and Chinese math problems through CoT and TIR.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-0.6B-Base","name":"Qwen3-0.6B-Base","created_by":"qwen","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/Qwen/Qwen3-0.6B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","parameters":0.6},{"id":"Qwen/Qwen3-0.6B","name":"Qwen3-0.6B","created_by":"qwen","source":"official","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","parameters":0.75,"page_url":"https://huggingface.co/Qwen/Qwen3-0.6B","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-1.7B-Base","name":"Qwen3-1.7B-Base","created_by":"qwen","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","parameters":1.7,"page_url":"https://huggingface.co/Qwen/Qwen3-1.7B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-1.7B","name":"Qwen3-1.7B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":2,"page_url":"https://huggingface.co/Qwen/Qwen3-1.7B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-14B-Base","name":"Qwen3-14B-Base","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-14B-Base","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-14B","name":"Qwen3-14B","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-14B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Instruct-2507","name":"Qwen3-235B-A22B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","alias":"Qwen/Qwen3-235B-A22B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507","release_date":"2025-07-21","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Instruct","name":"Qwen3-235B-A22B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"snapshots":["Qwen/Qwen3-235B-A22B-Instruct-2507"],"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507","release_date":"2025-07-21","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Thinking-2507","name":"Qwen3-235B-A22B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","alias":"Qwen/Qwen3-235B-A22B-Thinking","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507","release_date":"2025-07-25","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Thinking","name":"Qwen3-235B-A22B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"snapshots":["Qwen/Qwen3-235B-A22B-Thinking-2507"],"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507","release_date":"2025-07-25","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B","name":"Qwen3-235B-A22B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-30B-A3B-Instruct-2507","name":"Qwen3-30B-A3B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","alias":"Qwen/Qwen3-30B-A3B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507","release_date":"2025-07-28","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-30B-A3B-Instruct","name":"Qwen3-30B-A3B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"snapshots":["Qwen/Qwen3-30B-A3B-Instruct-2507"],"model_type":"chat","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507","release_date":"2025-07-28","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-30B-A3B-Thinking-2507","name":"Qwen3-30B-A3B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","release_date":"2025-07-29","context_window":262144,"model_type":"chat","license":"apache-2.0","parameters":31,"active_parameters":1.9,"alias":"Qwen/Qwen3-30B-A3B-Thinking","page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507","architecture":"moe","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"tools":["function_calling"]},{"id":"Qwen/Qwen3-30B-A3B-Thinking","name":"Qwen3-30B-A3B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","release_date":"2025-07-29","context_window":262144,"license":"apache-2.0","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"snapshots":["Qwen/Qwen3-30B-A3B-Thinking-2507"]},{"id":"Qwen/Qwen3-30B-A3B","name":"Qwen3-30B-A3B","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-32B","name":"Qwen3-32B","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen3-32B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-Base","name":"Qwen3-4B-Base","created_by":"qwen","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-4B-Instruct-2507","name":"Qwen3-4B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","alias":"Qwen/Qwen3-4B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-Instruct","name":"Qwen3-4B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["Qwen/Qwen3-4B-Instruct-2507"],"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-SafeRL","name":"Qwen3-4B-SafeRL","created_by":"qwen","source":"official","release_date":"2025-09-30","context_window":40960,"license":"apache-2.0","parameters":4.4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-SafeRL","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-4B-Thinking-2507","name":"Qwen3-4B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","alias":"Qwen/Qwen3-4B-Thinking","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-Thinking","name":"Qwen3-4B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["Qwen/Qwen3-4B-Thinking-2507"],"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B","name":"Qwen3-4B","created_by":"qwen","source":"official","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/Qwen/Qwen3-4B","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","parameters":4},{"id":"Qwen/Qwen3-8B-Base","name":"Qwen3-8B-Base","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8.2,"page_url":"https://huggingface.co/Qwen/Qwen3-8B-Base","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-8B","name":"Qwen3-8B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":8.2,"page_url":"https://huggingface.co/Qwen/Qwen3-8B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-30B-A3B-Instruct","name":"Qwen3-Coder-30B-A3B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct","release_date":"2025-07-31","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8","name":"Qwen3-Coder-480B-A35B-Instruct-FP8","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":480,"active_parameters":24,"release_date":"2025-07-22","context_window":262144,"license":"apache-2.0","page_url":"https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-480B-A35B-Instruct","name":"Qwen3-Coder-480B-A35B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","parameters":480,"active_parameters":24,"page_url":"https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct","release_date":"2025-07-22","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-Next-FP8","name":"Qwen3-Coder-Next-FP8","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","page_url":"https://huggingface.co/Qwen/Qwen3-Coder-Next-FP8","release_date":"2026-02-01","context_window":262144,"license":"apache-2.0","parameters":80,"active_parameters":1.6,"architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-Next","name":"Qwen3-Coder-Next","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","release_date":"2026-01-30","context_window":262144,"license":"apache-2.0","parameters":80,"active_parameters":1.6,"page_url":"https://huggingface.co/Qwen/Qwen3-Coder-Next","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","max_output_tokens":65536},{"id":"Qwen/Qwen3-Next-80B-A3B-Instruct","name":"Qwen3-Next-80B-A3B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":81,"active_parameters":1.6,"page_url":"https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct","description":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI). We are excited to share our latest advancements in addressing these demands, centered on improving scaling efficiency through innovative model architecture. We call this next-generation foundation models **Qwen3-Next**.","release_date":"2025-09-09","context_window":262144,"license":"apache-2.0","tagline":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI).","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Next-80B-A3B-Thinking","name":"Qwen3-Next-80B-A3B-Thinking","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":81,"active_parameters":1.6,"page_url":"https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking","description":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI). We are excited to share our latest advancements in addressing these demands, centered on improving scaling efficiency through innovative model architecture. We call this next-generation foundation models **Qwen3-Next**.","release_date":"2025-09-09","context_window":262144,"license":"apache-2.0","tagline":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI).","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3Guard-Gen-0.6B","name":"Qwen3Guard-Gen-0.6B","created_by":"qwen","source":"official","description":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: **Qwen3Guard-Gen**, a generative model that frames safety classification as an instruction-following task, and **Qwen3Guard-Stream**, which incorporates a token-level classification head for real-time safety monitoring during incremental text ge…","release_date":"2025-09-23","context_window":32768,"license":"apache-2.0","parameters":0.75,"page_url":"https://huggingface.co/Qwen/Qwen3Guard-Gen-0.6B","architecture":"transformer","model_type":"chat","tagline":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3Guard-Gen-4B","name":"Qwen3Guard-Gen-4B","created_by":"qwen","source":"official","description":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: **Qwen3Guard-Gen**, a generative model that frames safety classification as an instruction-following task, and **Qwen3Guard-Stream**, which incorporates a token-level classification head for real-time safety monitoring during incremental text ge…","release_date":"2025-09-23","context_window":32768,"license":"apache-2.0","parameters":4.4,"page_url":"https://huggingface.co/Qwen/Qwen3Guard-Gen-4B","architecture":"transformer","model_type":"chat","tagline":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"Qwen/Qwen3Guard-Gen-8B","name":"Qwen3Guard-Gen-8B","created_by":"qwen","source":"official","description":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: **Qwen3Guard-Gen**, a generative model that frames safety classification as an instruction-following task, and **Qwen3Guard-Stream**, which incorporates a token-level classification head for real-time safety monitoring during incremental text ge…","release_date":"2025-09-23","context_window":32768,"license":"apache-2.0","parameters":8.2,"page_url":"https://huggingface.co/Qwen/Qwen3Guard-Gen-8B","architecture":"transformer","model_type":"chat","tagline":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV","name":"Meta-Llama-3-8B-Instruct-FP8-KV","created_by":"redhatai","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV","release_date":"2024-05-20","context_window":8192,"architecture":"transformer","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"RedHatAI/Qwen2-1.5B-Instruct-FP8","name":"Qwen2-1.5B-Instruct-FP8","created_by":"redhatai","source":"official","release_date":"2024-06-14","context_window":32768,"license":"apache-2.0","parameters":1.5,"page_url":"https://huggingface.co/RedHatAI/Qwen2-1.5B-Instruct-FP8","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"RedHatAI/Qwen2-7B-Instruct-FP8","name":"Qwen2-7B-Instruct-FP8","created_by":"redhatai","source":"official","release_date":"2024-06-14","context_window":32768,"license":"apache-2.0","parameters":7.6,"page_url":"https://huggingface.co/RedHatAI/Qwen2-7B-Instruct-FP8","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-14"},{"id":"ricdomolm/mini-coder-1.7b","name":"mini-coder-1.7b","created_by":"ricdomolm","source":"official","description":"`mini-coder-1.7b` is a 1.7B parameter model distilled from Qwen 3 Coder 30B A3B. It punches well above its weight, outperforming SWE-agent-LM 7B on [SWE-bench Verified Bash only](https://www.swebench.com/):","release_date":"2025-09-30","context_window":40960,"license":"mit","parameters":1.7,"page_url":"https://huggingface.co/ricdomolm/mini-coder-1.7b","architecture":"transformer","tagline":"`mini-coder-1.7b` is a 1.7B parameter model distilled from Qwen 3 Coder 30B A3B.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"SakanaAI/TinySwallow-1.5B-Instruct","name":"TinySwallow-1.5B-Instruct","created_by":"sakanaai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":1.5,"page_url":"https://huggingface.co/SakanaAI/TinySwallow-1.5B-Instruct","description":"🤗 [Models](https://huggingface.co/SakanaAI) | 📚 [Paper](https://arxiv.org/abs/2501.16937) | 📝 [Blog](https://sakana.ai/taid-jp/) | 🐦 [Twitter](https://twitter.com/SakanaAILabs)","release_date":"2025-01-07","context_window":32768,"license":"apache-2.0","tagline":"🤗 [Models](https://huggingface.co/SakanaAI) | 📚 [Paper](https://arxiv.org/abs/2501.16937) | 📝 [Blog](https://sakana.ai/taid-jp/) | 🐦 [Twitter](https://twitter.com/SakanaAILabs)","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"ShahriarFerdoush/llama-3.2-1b-code-instruct","name":"llama-3.2-1b-code-instruct","created_by":"shahriarferdoush","source":"official","family":"llama-3.2","description":"A lightweight yet powerful **code-focused language model** fine-tuned from **Meta Llama-3.2-1B** using **QLoRA (4-bit)** on the **CodeAlpaca-20K** dataset. Designed for **efficient code generation, reasoning, and problem-solving** on limited GPU resources.","release_date":"2025-12-24","context_window":131072,"license":"apache-2.0","parameters":1.2,"page_url":"https://huggingface.co/ShahriarFerdoush/llama-3.2-1b-code-instruct","architecture":"transformer","model_type":"chat","tagline":"A lightweight yet powerful **code-focused language model** fine-tuned from **Meta Llama-3.2-1B** using **QLoRA (4-bit)** on the **CodeAlpaca-20K** dataset.","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-07"},{"id":"speakleash/Bielik-7B-Instruct-v0.1","name":"Bielik-7B-Instruct-v0.1","created_by":"speakleash","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"swap-uniba/LLaMAntino-3-ANITA-8B-Inst-DPO-ITA","name":"LLaMAntino-3-ANITA-8B-Inst-DPO-ITA","created_by":"swap-uniba","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/swap-uniba/LLaMAntino-3-ANITA-8B-Inst-DPO-ITA","description":"📣 New MODEL FAMILY❗ [https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA](https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA)","release_date":"2024-04-29","context_window":8192,"license":"llama3","tagline":"📣 New MODEL FAMILY❗ [https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA](https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA)","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"swiss-ai/Apertus-70B-Instruct-2509","name":"Apertus-70B-Instruct-2509","created_by":"swiss-ai","source":"official","release_date":"2025-09-01","context_window":65536,"license":"apache-2.0","parameters":71,"alias":"swiss-ai/Apertus-70B-Instruct","page_url":"https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"swiss-ai/Apertus-70B-Instruct","name":"Apertus-70B-Instruct-2509","created_by":"swiss-ai","source":"official","release_date":"2025-09-01","context_window":65536,"license":"apache-2.0","parameters":71,"page_url":"https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"snapshots":["swiss-ai/Apertus-70B-Instruct-2509"],"last_updated":"2026-06-10"},{"id":"swiss-ai/Apertus-8B-Instruct-2509","name":"Apertus-8B-Instruct-2509","created_by":"swiss-ai","source":"official","last_updated":"2026-04-21","alias":"swiss-ai/Apertus-8B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":8.1,"page_url":"https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509","release_date":"2025-08-13","context_window":65536,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"swiss-ai/Apertus-8B-Instruct","name":"Apertus-8B-Instruct-2509","created_by":"swiss-ai","source":"official","last_updated":"2026-04-21","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["swiss-ai/Apertus-8B-Instruct-2509"],"parameters":8.1,"page_url":"https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509","release_date":"2025-08-13","context_window":65536,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"TinyLlama/TinyLlama-1.1B-Chat-v1.0","name":"TinyLlama-1.1B-Chat-v1.0","created_by":"tinyllama","source":"official","release_date":"2023-12-30","context_window":2048,"license":"apache-2.0","parameters":1.1,"page_url":"https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1","name":"Llama-3-Swallow-8B-Instruct-v0.1","created_by":"tokyotech-llm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1","description":"Our Swallow model has undergone continual pre-training from the [Llama 3 family](https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6), primarily with the addition of Japanese language data. The Instruct versions use supervised fine-tuning (SFT) and Chat Vector. Links to other models can be found in the index.","release_date":"2024-06-26","context_window":8192,"license":"llama3","tagline":"Our Swallow model has undergone continual pre-training from the [Llama 3 family](https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6), primarily with the addition of Japanese language data.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.5","name":"Llama-3.1-Swallow-8B-Instruct-v0.5","created_by":"tokyotech-llm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.5","description":"Llama 3.1 Swallow is a series of large language models (8B, 70B) that were built by continual pre-training on the [Meta Llama 3.1](https://huggingface.co/collections/meta-llama/llama-31-669fc079a0c406a149a5738f) models. Llama 3.1 Swallow enhanced the Japanese language capabilities of the original Llama 3.1 while retaining the English language capabilities.","release_date":"2025-06-12","context_window":131072,"license":"llama3.3","tagline":"Llama 3.1 Swallow is a series of large language models (8B, 70B) that were built by continual pre-training on the [Meta Llama 3.1](https://huggingface.co/collections/meta-llama/llama-31-669fc079a0c406a149a5738f) models.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/gemma-2-9b-it","name":"gemma-2-9b-it","created_by":"unsloth","source":"official","last_updated":"2026-03-27","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":9.2,"page_url":"https://huggingface.co/unsloth/gemma-2-9b-it","release_date":"2024-07-03","context_window":8192,"license":"gemma","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/gemma-3-1b-it","name":"gemma-3-1b-it","created_by":"unsloth","source":"official","family":"gemma-3","description":"- Fine-tune Gemma 3 (12B) for free using our Google [Colab notebook here](https://docs.unsloth.ai/get-started/unsloth-notebooks)! - Read our Blog about Gemma 3 support: [unsloth.ai/blog/gemma3](https://unsloth.ai/blog/gemma3) - View the rest of our notebooks in our [docs here](https://docs.unsloth.ai/get-started/unsloth-notebooks). - Export your fine-tuned model to GGUF, Ollama, llama.cpp or 🤗HF.","release_date":"2025-03-12","context_window":32768,"license":"gemma","parameters":1,"page_url":"https://huggingface.co/unsloth/gemma-3-1b-it","architecture":"transformer","model_type":"chat","tagline":"- Fine-tune Gemma 3 (12B) for free using our Google [Colab notebook here](https://docs.unsloth.ai/get-started/unsloth-notebooks)! - Read our Blog about Gemma 3 support: [unsloth.ai/blog/gemma3](https://unsloth.ai/blog/gemma3) - View the rest of our notebooks in our [docs here](https://docs.unsloth.ai/get-started/unsloth-notebooks).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/llama-3-8b-Instruct","name":"llama-3-8b-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","family":"llama-3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/llama-3-8b-Instruct","description":"Directly quantized 4bit model with `bitsandbytes`.","release_date":"2024-04-18","context_window":8192,"license":"llama3","tagline":"Directly quantized 4bit model with `bitsandbytes`.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Llama-3.1-8B-Instruct","name":"Llama-3.1-8B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/Llama-3.1-8B-Instruct","release_date":"2025-02-15","context_window":131072,"license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Llama-3.2-1B-Instruct","name":"Llama-3.2-1B-Instruct","created_by":"unsloth","source":"official","family":"Llama-3.2","release_date":"2024-09-25","context_window":131072,"license":"llama3.2","parameters":1.2,"page_url":"https://huggingface.co/unsloth/Llama-3.2-1B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Llama-3.2-3B-Instruct","name":"Llama-3.2-3B-Instruct","created_by":"unsloth","source":"official","family":"Llama-3.2","release_date":"2024-09-25","context_window":131072,"license":"llama3.2","parameters":3.2,"page_url":"https://huggingface.co/unsloth/Llama-3.2-3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Llama-3.3-70B-Instruct","name":"Llama-3.3-70B-Instruct","created_by":"unsloth","source":"official","family":"Llama-3.3","release_date":"2024-12-06","context_window":131072,"license":"llama3.3","parameters":71,"page_url":"https://huggingface.co/unsloth/Llama-3.3-70B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-06"},{"id":"unsloth/Meta-Llama-3.1-8B-Instruct","name":"Meta-Llama-3.1-8B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/Meta-Llama-3.1-8B-Instruct","description":"We have a free Google Colab Tesla T4 notebook for Llama 3.1 (8B) here: https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing","release_date":"2024-07-23","context_window":131072,"license":"llama3.1","tagline":"We have a free Google Colab Tesla T4 notebook for Llama 3.1 (8B) here: https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Mistral-Small-24B-Instruct-2501","name":"Mistral-Small-24B-Instruct-2501","created_by":"unsloth","source":"official","last_updated":"2026-03-24","alias":"unsloth/Mistral-Small-24B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":24,"page_url":"https://huggingface.co/unsloth/Mistral-Small-24B-Instruct-2501","description":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","release_date":"2025-01-30","context_window":32768,"license":"apache-2.0","tagline":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Mistral-Small-24B-Instruct","name":"Mistral-Small-24B-Instruct-2501","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["unsloth/Mistral-Small-24B-Instruct-2501"],"model_type":"chat","parameters":24,"page_url":"https://huggingface.co/unsloth/Mistral-Small-24B-Instruct-2501","description":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","release_date":"2025-01-30","context_window":32768,"license":"apache-2.0","tagline":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-0.5B-Instruct","name":"Qwen2.5-0.5B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/unsloth/Qwen2.5-0.5B-Instruct","architecture":"transformer","model_type":"chat","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen2.5-1.5B-Instruct","name":"Qwen2.5-1.5B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","parameters":1.5,"page_url":"https://huggingface.co/unsloth/Qwen2.5-1.5B-Instruct","architecture":"transformer","model_type":"chat","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen2.5-14B-Instruct","name":"Qwen2.5-14B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/unsloth/Qwen2.5-14B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-23","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-32B-Instruct","name":"Qwen2.5-32B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":33,"page_url":"https://huggingface.co/unsloth/Qwen2.5-32B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-23","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-3B-Instruct","name":"Qwen2.5-3B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"other","parameters":3.1,"page_url":"https://huggingface.co/unsloth/Qwen2.5-3B-Instruct","architecture":"transformer","model_type":"chat","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen2.5-7B-Instruct","name":"Qwen2.5-7B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/unsloth/Qwen2.5-7B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-Coder-32B-Instruct","name":"Qwen2.5-Coder-32B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":33,"page_url":"https://huggingface.co/unsloth/Qwen2.5-Coder-32B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-11-12","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-Coder-7B-Instruct","name":"Qwen2.5-Coder-7B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-23","context_window":32768,"license":"apache-2.0","parameters":7.6,"page_url":"https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct","architecture":"transformer","model_type":"code","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-16"},{"id":"unsloth/Qwen3-1.7B-Base","name":"Qwen3-1.7B-Base","created_by":"unsloth","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-1.7B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-17"},{"id":"unsloth/Qwen3-4B-Base","name":"Qwen3-4B-Base","created_by":"unsloth","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-4B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02"},{"id":"unsloth/Qwen3-4B-Instruct-2507","name":"Qwen3-4B-Instruct-2507","created_by":"unsloth","source":"official","release_date":"2025-08-06","context_window":262144,"model_type":"chat","license":"apache-2.0","alias":"unsloth/Qwen3-4B-Instruct","page_url":"https://huggingface.co/unsloth/Qwen3-4B-Instruct-2507","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen3-4B-Instruct","name":"Qwen3-4B-Instruct-2507","created_by":"unsloth","source":"official","release_date":"2025-08-06","context_window":262144,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-4B-Instruct-2507","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"snapshots":["unsloth/Qwen3-4B-Instruct-2507"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen3-8B-Base","name":"Qwen3-8B-Base","created_by":"unsloth","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-8B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-03-26"},{"id":"unsloth/Qwen3-8B","name":"Qwen3-8B","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/Qwen3-8B","release_date":"2025-04-28","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/tinyllama-chat","name":"tinyllama-chat","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"page_url":"https://huggingface.co/unsloth/tinyllama-chat","description":"[![](\"https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord%20button.png\")
](https://discord.gg/u54VK8m8tk) [![](\"https://raw.githubusercontent.com/unslothai/unsloth/main/images/buy%20me%20a%20coffee%20button.png\")
](https://ko-fi.com/unsloth) [![](\"https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png\")
](https://github.com/unslothai/unsloth)","release_date":"2024-02-14","context_window":2048,"license":"apache-2.0","parameters":1.1,"tagline":"[](https://discord.gg/u54VK8m8tk) [![](\"https://raw.githubusercontent.com/unslothai/un","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"upstage/SOLAR-10.7B-Instruct-v1.0","name":"SOLAR-10.7B-Instruct-v1.0","created_by":"upstage","source":"official","release_date":"2023-12-12","context_window":4096,"license":"cc-by-nc-4.0","parameters":11,"page_url":"https://huggingface.co/upstage/SOLAR-10.7B-Instruct-v1.0","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"utter-project/EuroLLM-22B-Instruct-2512","name":"EuroLLM-22B-Instruct-2512","created_by":"utter-project","source":"official","description":"This)
Discord community.","release_date":"2025-07-20","context_window":131072,"license":"mit","tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.5-Air","name":"GLM-4.5-Air","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.5","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.5-Air","description":"👋 Join our Discord community.","release_date":"2025-07-20","context_window":131072,"license":"mit","parameters":110,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.5","name":"GLM-4.5","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.5","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.5","description":"👋 Join our Discord community.","release_date":"2025-07-20","context_window":131072,"license":"mit","parameters":358,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.6-FP8","name":"GLM-4.6-FP8","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.6","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.6-FP8","description":"👋 Join our Discord community.","release_date":"2025-09-29","context_window":202752,"license":"mit","tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.6","name":"GLM-4.6","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.6","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.6","description":"👋 Join our Discord community.","release_date":"2025-09-29","context_window":202752,"license":"mit","parameters":357,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.7-Flash","name":"GLM-4.7-Flash","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.7","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.7-Flash","description":"👋 Join our Discord community.","release_date":"2026-01-19","context_window":202752,"license":"mit","parameters":31,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.7-FP8","name":"GLM-4.7-FP8","created_by":"zhipu","source":"official","last_updated":"2026-05-29","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.7","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.7-FP8","description":"👋 Join our Discord community.","release_date":"2025-12-22","context_window":202752,"license":"mit","tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"parameters":358},{"id":"zai-org/GLM-4.7","name":"GLM-4.7","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.7","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.7","description":"👋 Join our Discord community.","release_date":"2025-12-22","context_window":202752,"license":"mit","parameters":358,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-5.1-FP8","name":"GLM-5.1-FP8","created_by":"zhipu","source":"official","family":"GLM-5.1","release_date":"2026-04-03","context_window":202752,"license":"mit","parameters":754,"page_url":"https://huggingface.co/zai-org/GLM-5.1-FP8","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-10"},{"id":"zai-org/GLM-5.1","name":"GLM-5.1","created_by":"zhipu","source":"official","family":"GLM-5.1","release_date":"2026-04-03","context_window":202752,"license":"mit","parameters":754,"page_url":"https://huggingface.co/zai-org/GLM-5.1","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-12"},{"id":"zai-org/GLM-5.2-FP8","name":"GLM-5.2-FP8","created_by":"zhipu","source":"official","family":"GLM-5.2","release_date":"2026-06-16","context_window":1048576,"license":"mit","parameters":753,"page_url":"https://huggingface.co/zai-org/GLM-5.2-FP8","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"zai-org/GLM-5","name":"GLM-5","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-5","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-5","description":"👋 Join our WeChat or Discord community.","release_date":"2026-02-11","context_window":202752,"license":"mit","parameters":754,"tagline":"👋 Join our WeChat or Discord","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-Z1-32B-0414","name":"GLM-Z1-32B-0414","created_by":"zhipu","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"chat","family":"glm","parameters":32}]};
+export const provider: ProviderWithModels = {"id":"huggingface","name":"Hugging Face","region":"US","headquarters":"New York, NY","founded":2016,"url":"https://huggingface.co","api_url":"https://router.huggingface.co/v1","docs_url":"https://huggingface.co/docs/api-inference","pricing_url":"https://huggingface.co/pricing","description":"The AI community platform hosting open-source models with serverless inference endpoints.","type":"aggregator","sdk":{"python":"huggingface_hub","javascript":"@huggingface/inference"},"openai_compatible":true,"free_tier":true,"github_url":"https://github.com/huggingface","models_url":"https://huggingface.co/models","twitter_url":"https://x.com/huggingface","discord_url":"https://discord.com/invite/hugging-face-879548962464493619","blog_url":"https://huggingface.co/blog","terms_url":"https://huggingface.co/terms-of-service","support_url":"https://huggingface.co/support","icon":"","models":[{"id":"0xgr3y/Qwen2.5-Coder-0.5B-Instruct-Gensyn-Swarm-tall_tame_panther","name":"Qwen2.5-Coder-0.5B-Instruct-Gensyn-Swarm-tall_tame_panther","created_by":"0xgr3y","source":"official","release_date":"2025-11-13","context_window":32768,"license":"mit","parameters":0.49,"page_url":"https://huggingface.co/0xgr3y/Qwen2.5-Coder-0.5B-Instruct-Gensyn-Swarm-tall_tame_panther","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"aaditya/Llama3-OpenBioLLM-8B","name":"Llama3-OpenBioLLM-8B","created_by":"aaditya","source":"official","last_updated":"2026-03-24","page_url":"https://huggingface.co/aaditya/Llama3-OpenBioLLM-8B","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"description":"Introducing OpenBioLLM-8B: A State-of-the-Art Open Source Biomedical Large Language Model","release_date":"2024-04-20","context_window":8192,"license":"llama3","tagline":"Introducing OpenBioLLM-8B: A State-of-the-Art Open Source Biomedical Large Language Model","architecture":"transformer","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"allenai/Olmo-3-7B-Instruct","name":"Olmo-3-7B-Instruct","created_by":"allenai","source":"official","last_updated":"2026-06-29","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":7.3,"page_url":"https://huggingface.co/allenai/Olmo-3-7B-Instruct","release_date":"2025-11-19","context_window":65536,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"allenai/Olmo-3.1-32B-Instruct","name":"Olmo-3.1-32B-Instruct","created_by":"allenai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":32,"page_url":"https://huggingface.co/allenai/Olmo-3.1-32B-Instruct","release_date":"2025-12-10","context_window":65536,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"alpindale/Llama-3.2-1B-Instruct","name":"Llama-3.2-1B-Instruct","created_by":"alpindale","source":"official","family":"Llama-3.2","release_date":"2024-09-25","context_window":131072,"license":"llama3.2","parameters":1.2,"page_url":"https://huggingface.co/alpindale/Llama-3.2-1B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"anthracite-org/magnum-v4-72b","name":"magnum-v4-72b","created_by":"anthracite-org","source":"official","description":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus.","release_date":"2024-09-20","context_window":32768,"license":"apache-2.0","parameters":73,"page_url":"https://huggingface.co/anthracite-org/magnum-v4-72b","architecture":"transformer","tagline":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"arcee-ai/Trinity-Large-Thinking","name":"Trinity-Large-Thinking","created_by":"arcee-ai","source":"official","description":"src=\"https://cdn-uploads.huggingface.co/production/uploads/6435718aaaef013d1aec3b8b/i-v1KyAMOW_mgVGeic9WJ.png\" alt=\"Arcee Trinity Large Thinking\" style=\"max-width: 100%; height: auto;\" >","release_date":"2026-04-01","context_window":262144,"license":"apache-2.0","parameters":399,"active_parameters":6.2,"page_url":"https://huggingface.co/arcee-ai/Trinity-Large-Thinking","architecture":"moe","tagline":"src=\"https://cdn-uploads.huggingface.co/production/uploads/6435718aaaef013d1aec3b8b/i-v1KyAMOW_mgVGeic9WJ.png\" alt=\"Arcee Trinity Large Thinking\" style=\"max-width: 100%; height: auto;\" >","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-10"},{"id":"argilla/Llama-3.2-1B-Instruct-APIGen-FC-v0.1","name":"Llama-3.2-1B-Instruct-APIGen-FC-v0.1","created_by":"argilla","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.2","model_type":"chat","parameters":1,"page_url":"https://huggingface.co/argilla/Llama-3.2-1B-Instruct-APIGen-FC-v0.1","description":"This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on [argilla-warehouse/apigen-synth-trl](https://huggingface.co/datasets/argilla-warehouse/apigen-synth-trl) dataset, a version of [argilla/Synth-APIGen-v0.1](https://huggingface.co/datasets/argilla-warehouse/Synth-APIGen-v0.1) ready to do SFT on top of it. It has been trained using [TRL](https://github.com/huggingface/trl).","release_date":"2024-10-07","context_window":131072,"license":"apache-2.0","tagline":"This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on [argilla-warehouse/apigen-synth-trl](https://huggingface.co/datasets/argilla-warehouse/apigen-synth-trl) dataset, a version of [argilla/Synth-APIGen-v0.1](https://huggingface.co/datasets/argilla-warehouse/Synth-APIGen-v0.1) ready to do SFT on top of it.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"argilla-warehouse/Llama-3.2-1B-Instruct-v2-FC","name":"Llama-3.2-1B-Instruct-v2-FC","created_by":"argilla-warehouse","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"Llama-3.2","model_type":"chat","parameters":1,"page_url":"https://huggingface.co/argilla-warehouse/Llama-3.2-1B-Instruct-v2-FC","description":"This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on the [argilla-warehouse/apigen-smollm-trl-FC](https://huggingface.co/datasets/argilla-warehouse/apigen-smollm-trl-FC) dataset. It has been trained using [TRL](https://github.com/huggingface/trl).","release_date":"2024-10-20","context_window":131072,"tagline":"This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on the [argilla-warehouse/apigen-smollm-trl-FC](https://huggingface.co/datasets/argilla-warehouse/apigen-smollm-trl-FC) dataset.","architecture":"transformer","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"baseten/Llama-3.2-3B-Instruct-pythonic","name":"Llama-3.2-3B-Instruct-pythonic","created_by":"baseten","source":"official","family":"Llama-3.2","release_date":"2025-09-12","context_window":131072,"license":"llama3.2","parameters":3.2,"page_url":"https://huggingface.co/baseten/Llama-3.2-3B-Instruct-pythonic","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"codellama/CodeLlama-7b-Instruct-hf","name":"CodeLlama-7b-Instruct-hf","created_by":"codellama","source":"official","family":"codellama","description":"Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 34 billion parameters. This is the repository for the 7B instruct-tuned version in the Hugging Face Transformers format. This model is designed for general code synthesis and understanding. Links to other models can be found in the index at the bottom.","release_date":"2023-08-24","context_window":16384,"license":"llama2","parameters":6.7,"page_url":"https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf","architecture":"transformer","model_type":"code","tagline":"Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 34 billion parameters.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"deepcogito/cogito-v1-preview-qwen-32B","name":"cogito-v1-preview-qwen-32B","created_by":"deepcogito","source":"official","release_date":"2025-03-31","context_window":131072,"license":"apache-2.0","page_url":"https://huggingface.co/deepcogito/cogito-v1-preview-qwen-32B","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","parameters":33},{"id":"DeepMount00/Llama-3-8b-Ita","name":"Llama-3-8b-Ita","created_by":"deepmount00","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/DeepMount00/Llama-3-8b-Ita","description":"**💡 Found this resource helpful?** Creating and maintaining open source AI models and datasets requires significant computational resources. If this work has been valuable to you, consider [supporting my research](https://buymeacoffee.com/michele.montebovi) to help me continue building tools that benefit the entire AI community. Every contribution directly funds more open source innovation! ☕","release_date":"2024-05-01","context_window":8192,"license":"llama3","tagline":"**💡 Found this resource helpful?** Creating and maintaining open source AI models and datasets requires significant computational resources.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/DeepSeek-R1-0528-Qwen3-8B","name":"DeepSeek-R1-0528-Qwen3-8B","created_by":"deepseek","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"reasoning","parameters":8.2,"page_url":"https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B","release_date":"2025-05-29","context_window":131072,"license":"mit","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"defog/sqlcoder-7b-2","name":"sqlcoder-7b-2","created_by":"defog","source":"official","description":"The model weights were updated at 7 AM UTC on Feb 7, 2024. The new model weights lead to a much more performant model – particularly for joins.","release_date":"2024-02-05","context_window":16384,"license":"cc-by-sa-4.0","parameters":6.7,"page_url":"https://huggingface.co/defog/sqlcoder-7b-2","architecture":"transformer","tagline":"The model weights were updated at 7 AM UTC on Feb 7, 2024.","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"last_updated":"2026-06-18"},{"id":"DeSTA-ntu/Llama-3.1-8B-Instruct","name":"Llama-3.1-8B-Instruct","created_by":"desta-ntu","source":"official","family":"Llama-3.1","release_date":"2025-07-06","context_window":131072,"parameters":8,"page_url":"https://huggingface.co/DeSTA-ntu/Llama-3.1-8B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"dicta-il/DictaLM-3.0-1.7B-Instruct","name":"DictaLM-3.0-1.7B-Instruct","created_by":"dicta-il","source":"official","description":"[![](\"https://i.ibb.co/5Lbwyr1/dicta-logo.jpg\")
](https://dicta.org.il)","release_date":"2025-12-01","context_window":62080,"license":"apache-2.0","page_url":"https://huggingface.co/dicta-il/DictaLM-3.0-1.7B-Instruct","architecture":"transformer","tagline":"[](https://dicta.org.il)","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"dphn/Dolphin-Mistral-24B-Venice-Edition","name":"Dolphin-Mistral-24B-Venice-Edition","created_by":"dphn","source":"official","description":"Website: https://dphn.ai Twitter: https://x.com/dphnAI Web Chat: https://chat.dphn.ai Telegram bot: https://t.me/DolphinAI_bot","release_date":"2025-06-12","context_window":32768,"license":"apache-2.0","parameters":24,"page_url":"https://huggingface.co/dphn/Dolphin-Mistral-24B-Venice-Edition","architecture":"transformer","tagline":"Website: https://dphn.ai Twitter: https://x.com/dphnAI Web Chat: https://chat.dphn.ai Telegram bot: https://t.me/DolphinAI_bot","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-07"},{"id":"Efficient-Large-Model/gemma-2-2b-it","name":"gemma-2-2b-it","created_by":"efficient-large-model","source":"official","last_updated":"2026-03-24","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":2.6,"page_url":"https://huggingface.co/Efficient-Large-Model/gemma-2-2b-it","description":"**Model Page**: [Gemma](https://ai.google.dev/gemma/docs/base)","release_date":"2024-12-12","context_window":8192,"license":"gemma","tagline":"**Model Page**: [Gemma](https://ai.google.dev/gemma/docs/base)","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"EssentialAI/rnj-1-instruct","name":"rnj-1-instruct","created_by":"essentialai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"page_url":"https://huggingface.co/EssentialAI/rnj-1-instruct","description":"Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models. These models perform well across a range of programming languages and boast strong agentic capabilities (e.g., inside agentic frameworks like mini-SWE-agent), while also excelling at tool-calling. They additionally exhibit strong capabilities in math and science.","release_date":"2025-12-04","context_window":32768,"license":"apache-2.0","parameters":8.3,"tagline":"style=\"vertical-align: middle;\" src=\"https://img.shields.io/badge/%F0%9F%8C%90%20Website-essential.ai-4b9fe1?color=4b9fe1&logoColor=white\"/>","architecture":"transformer","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"failspy/llama-3-70B-Instruct-abliterated","name":"llama-3-70B-Instruct-abliterated","created_by":"failspy","source":"official","family":"llama-3","description":"This is meta-llama/Llama-3-70B-Instruct with orthogonalized bfloat16 safetensor weights, generated with the methodology that was described in the preview paper/blog post: '[Refusal in LLMs is mediated by a single direction](https://www.alignmentforum.org/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction)' which I encourage you to read to understand more.","release_date":"2024-05-07","context_window":8192,"license":"llama3","page_url":"https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated","architecture":"transformer","model_type":"chat","tagline":"This is meta-llama/Llama-3-70B-Instruct with orthogonalized bfloat16 safetensor weights, generated with the methodology that was described in the preview paper/blog post: '[Refusal in LLMs is mediated","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-07"},{"id":"failspy/Meta-Llama-3-8B-Instruct-abliterated-v3","name":"Meta-Llama-3-8B-Instruct-abliterated-v3","created_by":"failspy","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/failspy/Meta-Llama-3-8B-Instruct-abliterated-v3","description":"[My Jupyter \"cookbook\" to replicate the methodology can be found here, refined library coming soon](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)","release_date":"2024-05-20","context_window":8192,"license":"llama3","tagline":"[My Jupyter \"cookbook\" to replicate the methodology can be found here, refined library coming soon](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"farbodtavakkoli/OTel-LLM-0.6B-IT","name":"OTel-LLM-0.6B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-0.6B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-0.6B-IT","architecture":"transformer","tagline":"**OTel-LLM-0.6B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-1.2B-IT","name":"OTel-LLM-1.2B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-1.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-03-10","context_window":128000,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-1.2B-IT","architecture":"transformer","tagline":"**OTel-LLM-1.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-1.7B-IT","name":"OTel-LLM-1.7B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-1.7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-1.7B-IT","architecture":"transformer","tagline":"**OTel-LLM-1.7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-12B-IT","name":"OTel-LLM-12B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-12B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-12B-IT","architecture":"transformer","tagline":"**OTel-LLM-12B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-22"},{"id":"farbodtavakkoli/OTel-LLM-14B-IT","name":"OTel-LLM-14B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-14B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-14B-IT","architecture":"transformer","tagline":"**OTel-LLM-14B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-1B-IT","name":"OTel-LLM-1B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-1B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":32768,"license":"apache-2.0","parameters":1,"page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-1B-IT","architecture":"transformer","tagline":"**OTel-LLM-1B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-20B-IT","name":"OTel-LLM-20B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-20B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-13","context_window":131072,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-20B-IT","architecture":"moe","tagline":"**OTel-LLM-20B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-270M-IT","name":"OTel-LLM-270M-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-270M-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":32768,"license":"apache-2.0","parameters":0.27,"page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-270M-IT","architecture":"transformer","tagline":"**OTel-LLM-270M-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-27B-IT","name":"OTel-LLM-27B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-27B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-27B-IT","architecture":"transformer","tagline":"**OTel-LLM-27B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-22"},{"id":"farbodtavakkoli/OTel-LLM-32B-IT","name":"OTel-LLM-32B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-32B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":65536,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-32B-IT","architecture":"transformer","tagline":"**OTel-LLM-32B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-4B-IT","name":"OTel-LLM-4B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-4B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","license":"apache-2.0","parameters":4.3,"page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-4B-IT","architecture":"transformer","tagline":"**OTel-LLM-4B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-7B-IT","name":"OTel-LLM-7B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":65536,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-7B-IT","architecture":"transformer","tagline":"**OTel-LLM-7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-8.2B-IT","name":"OTel-LLM-8.2B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-8.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-8.2B-IT","architecture":"transformer","tagline":"**OTel-LLM-8.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-8.3B-IT","name":"OTel-LLM-8.3B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-8.3B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-8.3B-IT","architecture":"transformer","tagline":"**OTel-LLM-8.3B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"fdtn-ai/Foundation-Sec-8B-Instruct","name":"Foundation-Sec-8B-Instruct","created_by":"fdtn-ai","source":"official","release_date":"2025-07-29","context_window":131072,"license":"other","parameters":8,"page_url":"https://huggingface.co/fdtn-ai/Foundation-Sec-8B-Instruct","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"Featherless-Chat-Models/Mistral-7B-Instruct-v0.2","name":"Mistral-7B-Instruct-v0.2","created_by":"featherless-chat-models","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.2,"page_url":"https://huggingface.co/Featherless-Chat-Models/Mistral-7B-Instruct-v0.2","release_date":"2025-05-08","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"FlagAlpha/Llama3-Chinese-8B-Instruct","name":"Llama3-Chinese-8B-Instruct","created_by":"flagalpha","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/FlagAlpha/Llama3-Chinese-8B-Instruct","description":"Llama3-Chinese-8B-Instruct基于Llama3-8B中文微调对话模型，由Llama中文社区和AtomEcho（原子回声）联合研发，我们会持续提供更新的模型参数，模型训练过程见 [https://llama.family](https://llama.family)。","release_date":"2024-04-23","context_window":8192,"license":"apache-2.0","tagline":"Llama3-Chinese-8B-Instruct基于Llama3-8B中文微调对话模型，由Llama中文社区和AtomEcho（原子回声）联合研发，我们会持续提供更新的模型参数，模型训练过程见 [https://llama.family](https://llama.family)。","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-1.1-2b-it","name":"gemma-1.1-2b-it","created_by":"google","source":"official","family":"gemma-1","release_date":"2024-03-26","license":"gemma","parameters":2.5,"page_url":"https://huggingface.co/google/gemma-1.1-2b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"google/gemma-2-27b-it","name":"gemma-2-27b-it","created_by":"google","source":"official","family":"gemma-2","release_date":"2024-06-24","license":"gemma","parameters":27,"page_url":"https://huggingface.co/google/gemma-2-27b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"google/gemma-2-2b-it","name":"gemma-2-2b-it","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","release_date":"2024-07-16","license":"gemma","parameters":2.6,"page_url":"https://huggingface.co/google/gemma-2-2b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"google/gemma-2-2b-jpn-it","name":"gemma-2-2b-jpn-it","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":2.6,"page_url":"https://huggingface.co/google/gemma-2-2b-jpn-it","release_date":"2024-09-25","license":"gemma","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2-9b-it","name":"gemma-2-9b-it","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":9.2,"page_url":"https://huggingface.co/google/gemma-2-9b-it","release_date":"2024-06-24","license":"gemma","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2b-it","name":"gemma-2b-it","created_by":"google","source":"official","family":"gemma-2b","release_date":"2024-02-08","license":"gemma","page_url":"https://huggingface.co/google/gemma-2b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","parameters":2.5},{"id":"google/gemma-3-1b-it","name":"gemma-3-1b-it","created_by":"google","source":"official","family":"gemma-3","release_date":"2025-03-10","license":"gemma","parameters":1,"page_url":"https://huggingface.co/google/gemma-3-1b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"google/gemma-7b-it","name":"gemma-7b-it","created_by":"google","source":"official","family":"gemma-7b","release_date":"2024-02-13","license":"gemma","page_url":"https://huggingface.co/google/gemma-7b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18","parameters":8.5},{"id":"google/medgemma-27b-text-it","name":"medgemma-27b-text-it","created_by":"google","source":"official","release_date":"2025-05-19","license":"other","page_url":"https://huggingface.co/google/medgemma-27b-text-it","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-29","parameters":27},{"id":"GraySwanAI/Llama-3-8B-Instruct-RR","name":"Llama-3-8B-Instruct-RR","created_by":"grayswanai","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/GraySwanAI/Llama-3-8B-Instruct-RR","description":"Llama-3-8B-Instruct-RR is a Llama-3 model with circuit breakers inserted using Representation Rerouting (RR).","release_date":"2024-07-08","context_window":8192,"tagline":"Llama-3-8B-Instruct-RR is a Llama-3 model with circuit breakers inserted using Representation Rerouting (RR).","architecture":"transformer","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"GritLM/GritLM-7B","name":"GritLM-7B","created_by":"gritlm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":7.2,"page_url":"https://huggingface.co/GritLM/GritLM-7B","description":"> GritLM is a generative representational instruction tuned language model. It unifies text representation (embedding) and text generation into a single model achieving state-of-the-art performance on both types of tasks.","release_date":"2024-02-11","context_window":32768,"license":"apache-2.0","tagline":"> GritLM is a generative representational instruction tuned language model.","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"h2oai/h2ogpt-4096-llama2-7b-chat","name":"h2ogpt-4096-llama2-7b-chat","created_by":"h2oai","source":"official","description":"h2oGPT clone of [Meta's Llama 2 7B Chat](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf).","release_date":"2023-08-09","context_window":4096,"license":"llama2","parameters":6.7,"page_url":"https://huggingface.co/h2oai/h2ogpt-4096-llama2-7b-chat","architecture":"transformer","tagline":"h2oGPT clone of [Meta's Llama 2 7B Chat](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf).","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"last_updated":"2026-05-17"},{"id":"huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated","name":"DeepSeek-R1-Distill-Qwen-32B-abliterated","created_by":"huihui-ai","source":"official","description":"This is an uncensored version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.","release_date":"2025-01-22","context_window":131072,"parameters":33,"page_url":"https://huggingface.co/huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated","architecture":"transformer","model_type":"reasoning","tagline":"This is an uncensored version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it).","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-14"},{"id":"huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2","name":"Qwen2.5-14B-Instruct-abliterated-v2","created_by":"huihui-ai","source":"official","description":"This is an uncensored version of [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","release_date":"2024-10-09","context_window":32768,"license":"apache-2.0","parameters":15,"page_url":"https://huggingface.co/huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2","architecture":"transformer","model_type":"chat","tagline":"This is an uncensored version of [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteratio","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-21"},{"id":"huihui-ai/Qwen2.5-14B-Instruct-abliterated","name":"Qwen2.5-14B-Instruct-abliterated","created_by":"huihui-ai","source":"official","description":"This is an uncensored version of [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","release_date":"2024-09-25","context_window":32768,"license":"apache-2.0","parameters":15,"page_url":"https://huggingface.co/huihui-ai/Qwen2.5-14B-Instruct-abliterated","architecture":"transformer","model_type":"chat","tagline":"This is an uncensored version of [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/ablite","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"huihui-ai/Qwen2.5-72B-Instruct-abliterated","name":"Qwen2.5-72B-Instruct-abliterated","created_by":"huihui-ai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":73,"page_url":"https://huggingface.co/huihui-ai/Qwen2.5-72B-Instruct-abliterated","description":"This is an uncensored version of [Qwen/Qwen2.5-72B-Instruct](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.","release_date":"2024-10-26","context_window":32768,"license":"other","tagline":"This is an uncensored version of [Qwen/Qwen2.5-72B-Instruct](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"instruction-pretrain/finance-Llama3-8B","name":"finance-Llama3-8B","created_by":"instruction-pretrain","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"parameters":8,"page_url":"https://huggingface.co/instruction-pretrain/finance-Llama3-8B","description":"This repo contains the **finance model developed from Llama3-8B** in our paper [Instruction Pre-Training: Language Models are Supervised Multitask Learners](https://huggingface.co/papers/2406.14491).","release_date":"2024-06-18","context_window":8192,"license":"llama3","tagline":"This repo contains the **finance model developed from Llama3-8B** in our paper [Instruction Pre-Training: Language Models are Supervised Multitask Learners](https://huggingface.co/papers/2406.14491).","architecture":"transformer","open_weight":true},{"id":"Intel/neural-chat-7b-v3-3","name":"neural-chat-7b-v3-3","created_by":"intel","source":"official","last_updated":"2026-03-22","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":7.2,"page_url":"https://huggingface.co/Intel/neural-chat-7b-v3-3","release_date":"2023-12-09","context_window":32768,"license":"apache-2.0"},{"id":"invalid-coder/Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp","name":"Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp","created_by":"invalid-coder","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":11,"page_url":"https://huggingface.co/invalid-coder/Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp","description":"NeuralPipe-7B-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [jeonsworld/CarbonVillain-en-10.7B-v2](https://huggingface.co/jeonsworld/CarbonVillain-en-10.7B-v2) * [kyujinpy/Sakura-SOLAR-Instruct](https://huggingface.co/kyujinpy/Sakura-SOLAR-Instruct)","release_date":"2024-01-10","context_window":4096,"license":"apache-2.0","tagline":"NeuralPipe-7B-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [jeonsworld/CarbonVillain-en-10.7B","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"kakaocorp/kanana-1.5-8b-instruct-2505","name":"kanana-1.5-8b-instruct-2505","created_by":"kakaocorp","source":"official","release_date":"2025-05-21","context_window":32768,"license":"apache-2.0","alias":"kakaocorp/kanana-1.5-8b-instruct","page_url":"https://huggingface.co/kakaocorp/kanana-1.5-8b-instruct-2505","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"kakaocorp/kanana-1.5-8b-instruct","name":"kanana-1.5-8b-instruct-2505","created_by":"kakaocorp","source":"official","release_date":"2025-05-21","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/kakaocorp/kanana-1.5-8b-instruct-2505","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"snapshots":["kakaocorp/kanana-1.5-8b-instruct-2505"],"last_updated":"2026-06-10"},{"id":"llmfan46/Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic","name":"Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic","created_by":"llmfan46","source":"official","release_date":"2026-04-25","license":"apache-2.0","parameters":31,"page_url":"https://huggingface.co/llmfan46/Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"lmstudio-community/Qwen3-1.7B-MLX-bf16","name":"Qwen3-1.7B-MLX-bf16","created_by":"lmstudio-community","source":"official","description":"This model [lmstudio-community/Qwen3-1.7B-MLX-bf16](https://huggingface.co/lmstudio-community/Qwen3-1.7B-MLX-bf16) was converted to MLX format from [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) using mlx-lm version **0.24.0**.","release_date":"2025-04-28","context_window":40960,"license":"apache-2.0","parameters":1.7,"page_url":"https://huggingface.co/lmstudio-community/Qwen3-1.7B-MLX-bf16","architecture":"transformer","model_type":"chat","tagline":"This model [lmstudio-community/Qwen3-1.7B-MLX-bf16](https://huggingface.co/lmstudio-community/Qwen3-1.7B-MLX-bf16) was converted to MLX format from [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"m8than/Mistral-Nemo-Instruct-2407-lenient-chatfix","name":"Mistral-Nemo-Instruct-2407-lenient-chatfix","created_by":"m8than","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","page_url":"https://huggingface.co/m8than/Mistral-Nemo-Instruct-2407-lenient-chatfix","description":"It's just mistral nemo 2407 with a less strict chat format.","release_date":"2025-05-06","context_window":131072,"license":"apache-2.0","parameters":12,"tagline":"It's just mistral nemo 2407 with a less strict chat format.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta-llama/Llama-2-13b-chat-hf","name":"Llama-2-13b-chat-hf","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-2","model_type":"chat","parameters":13,"page_url":"https://huggingface.co/meta-llama/Llama-2-13b-chat-hf","release_date":"2023-07-13","license":"llama2","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":4096,"max_output_tokens":4096},{"id":"meta-llama/Llama-3.1-70B-Instruct","name":"Llama-3.1-70B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct","release_date":"2024-07-16","license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.1-8B-Instruct","name":"Llama-3.1-8B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct","release_date":"2024-07-18","license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.2-1B-Instruct","name":"Llama-3.2-1B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.2","model_type":"chat","parameters":1.2,"page_url":"https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct","release_date":"2024-09-18","license":"llama3.2","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.2-3B-Instruct","name":"Llama-3.2-3B-Instruct","created_by":"meta","source":"official","family":"Llama-3.2","release_date":"2024-09-18","license":"llama3.2","parameters":3.2,"page_url":"https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-14","context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.3-70B-Instruct","name":"Llama-3.3-70B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.3","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct","release_date":"2024-11-26","license":"llama3.3","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Meta-Llama-3-70B-Instruct","name":"Meta-Llama-3-70B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct","release_date":"2024-04-17","license":"llama3","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":8192,"max_output_tokens":8192},{"id":"meta-llama/Meta-Llama-3-8B-Instruct","name":"Meta-Llama-3-8B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct","release_date":"2024-04-17","license":"llama3","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":8192,"max_output_tokens":8192},{"id":"microsoft/Phi-3-mini-128k-instruct","name":"Phi-3-mini-128k-instruct","created_by":"microsoft","source":"official","description":"🎉**Phi-4**: [[multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx)]; [[mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-mini-instruct-onnx)]","release_date":"2024-04-22","context_window":131072,"license":"mit","parameters":3.8,"page_url":"https://huggingface.co/microsoft/Phi-3-mini-128k-instruct","architecture":"transformer","model_type":"chat","tagline":"🎉**Phi-4**: [[multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx)]; [[mini-instruct](https://hug","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"microsoft/Phi-3-mini-4k-instruct","name":"Phi-3-mini-4k-instruct","created_by":"microsoft","source":"official","description":"🎉 **Phi-3.5**: [[mini-instruct]](https://huggingface.co/microsoft/Phi-3.5-mini-instruct); [[MoE-instruct]](https://huggingface.co/microsoft/Phi-3.5-MoE-instruct) ; [[vision-instruct]](https://huggingface.co/microsoft/Phi-3.5-vision-instruct)","release_date":"2024-04-22","context_window":4096,"license":"mit","parameters":3.8,"page_url":"https://huggingface.co/microsoft/Phi-3-mini-4k-instruct","architecture":"transformer","model_type":"chat","tagline":"🎉 **Phi-3.5**: [[mini-instruct]](https://huggingface.co/microsoft/Phi-3.5-mini-instruct); [[MoE-instruct]](https://huggingface.co/microsoft/Phi-3.5-MoE-instruct) ; [[vision-instruct]](https://hugging","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"microsoft/Phi-4-mini-instruct","name":"Phi-4-mini-instruct","created_by":"microsoft","source":"official","description":"🎉**Phi-4**: [[mini-reasoning](https://huggingface.co/microsoft/Phi-4-mini-reasoning) | [reasoning](https://huggingface.co/microsoft/Phi-4-reasoning)] | [[multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx)]; [[mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-mini-instruct-onnx)]","release_date":"2025-02-19","context_window":131072,"license":"mit","page_url":"https://huggingface.co/microsoft/Phi-4-mini-instruct","architecture":"transformer","model_type":"chat","tagline":"🎉**Phi-4**: [[mini-reasoning](https://huggingface.co/microsoft/Phi-4-mini-reasoning) | [reasoning](https://huggingface.co/microsoft/Phi-4-reasoning)] | [[multimodal-instruct](https://huggingface.co/m","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","parameters":3.8},{"id":"mistralai/Mistral-7B-Instruct-v0.1","name":"Mistral-7B-Instruct-v0.1","created_by":"mistral","source":"official","release_date":"2023-09-27","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","parameters":7.2},{"id":"mistralai/Mistral-7B-Instruct-v0.2","name":"Mistral-7B-Instruct-v0.2","created_by":"mistral","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.2,"page_url":"https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2","release_date":"2023-12-11","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated","name":"Meta-Llama-3.1-8B-Instruct-abliterated","created_by":"mlabonne","source":"official","last_updated":"2026-03-24","family":"Llama-3.1","page_url":"https://huggingface.co/mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"This is an uncensored version of Llama 3.1 8B Instruct created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","release_date":"2024-07-24","context_window":131072,"license":"llama3.1","parameters":8,"tagline":"This is an uncensored version of Llama 3.1 8B Instruct created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mlabonne/NeuralDaredevil-8B-abliterated","name":"NeuralDaredevil-8B-abliterated","created_by":"mlabonne","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":8,"page_url":"https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated","description":"This is a DPO fine-tune of [mlabonne/Daredevil-8-abliterated](https://huggingface.co/mlabonne/Daredevil-8B-abliterated), trained on one epoch of [mlabonne/orpo-dpo-mix-40k](https://huggingface.co/datasets/mlabonne/orpo-dpo-mix-40k). The DPO fine-tuning successfully recovers the performance loss due to the abliteration process, making it an excellent uncensored model.","release_date":"2024-05-27","context_window":8192,"license":"llama3","tagline":"This is a DPO fine-tune of [mlabonne/Daredevil-8-abliterated](https://huggingface.co/mlabonne/Daredevil-8B-abliterated), trained on one epoch of [mlabonne/orpo-dpo-mix-40k](https://huggingface.co/datasets/mlabonne/orpo-dpo-mix-40k).","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"moonshotai/Kimi-K2-Instruct-0905","name":"Kimi-K2-Instruct-0905","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Kimi-K2","model_type":"chat","page_url":"https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905","release_date":"2025-09-03","context_window":262144,"license":"other","parameters":1026,"architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/Kimi-K2-Instruct","name":"Kimi-K2-Instruct","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Kimi-K2","model_type":"chat","page_url":"https://huggingface.co/moonshotai/Kimi-K2-Instruct","release_date":"2025-07-11","context_window":131072,"license":"other","parameters":1026,"architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/Kimi-K2-Thinking","name":"Kimi-K2-Thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true},"family":"Kimi-K2","model_type":"chat","page_url":"https://huggingface.co/moonshotai/Kimi-K2-Thinking","release_date":"2025-11-04","context_window":262144,"license":"other","parameters":1058,"architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Long-term thinking, multi-step tool usage, complex problem solving.","tagline":"Long-term thinking, multi-step tool usage, complex problem solving.","max_output_tokens":32768},{"id":"moonshotai/Kimi-Linear-48B-A3B-Instruct","name":"Kimi-Linear-48B-A3B-Instruct","created_by":"moonshot","source":"official","family":"kimi","release_date":"2025-10-30","license":"mit","parameters":49,"page_url":"https://huggingface.co/moonshotai/Kimi-Linear-48B-A3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"NousResearch/Meta-Llama-3-70B-Instruct","name":"Meta-Llama-3-70B-Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/NousResearch/Meta-Llama-3-70B-Instruct","release_date":"2024-04-19","context_window":8192,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"NousResearch/Meta-Llama-3-8B-Instruct","name":"Meta-Llama-3-8B-Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct","release_date":"2024-04-18","context_window":8192,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"NousResearch/Meta-Llama-3.1-8B-Instruct","name":"Meta-Llama-3.1-8B-Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/NousResearch/Meta-Llama-3.1-8B-Instruct","release_date":"2024-07-24","context_window":131072,"license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/Gemma-4-31B-IT-NVFP4","name":"Gemma-4-31B-IT-NVFP4","created_by":"nvidia","source":"official","release_date":"2026-04-02","license":"other","parameters":21,"page_url":"https://huggingface.co/nvidia/Gemma-4-31B-IT-NVFP4","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"nvidia/Llama-3.1-8B-Instruct-FP8","name":"Llama-3.1-8B-Instruct-FP8","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/nvidia/Llama-3.1-8B-Instruct-FP8","release_date":"2024-08-29","context_window":131072,"license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"structured_output":true,"reasoning":true,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":120,"page_url":"https://huggingface.co/openai/gpt-oss-120b","description":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","release_date":"2025-08-04","context_window":131072,"license":"apache-2.0","tagline":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":131072,"knowledge_cutoff":"2024-06"},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"structured_output":true,"reasoning":true,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":22,"page_url":"https://huggingface.co/openai/gpt-oss-20b","description":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","release_date":"2025-08-04","context_window":131072,"license":"apache-2.0","tagline":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":131072,"knowledge_cutoff":"2024-06"},{"id":"openai/gpt-oss-safeguard-20b","name":"gpt-oss-safeguard-20b","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"moderation","parameters":22,"page_url":"https://huggingface.co/openai/gpt-oss-safeguard-20b","description":"`gpt-oss-safeguard-120b` and `gpt-oss-safeguard-20b` are safety reasoning models built-upon gpt-oss. With these models, you can classify text content based on safety policies that you provide and perform a suite of foundational safety tasks. These models are intended for safety use cases. For other applications, we recommend using [gpt-oss models](https://huggingface.co/collections/openai/gpt-oss).","release_date":"2025-09-18","context_window":131072,"license":"apache-2.0","tagline":"`gpt-oss-safeguard-120b` and `gpt-oss-safeguard-20b` are safety reasoning models built-upon gpt-oss.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openchat/openchat-3.6-8b-20240522","name":"openchat-3.6-8b-20240522","created_by":"openchat","source":"official","last_updated":"2026-03-24","alias":"openchat/openchat-3.6-8b","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":8,"page_url":"https://huggingface.co/openchat/openchat-3.6-8b-20240522","description":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below. The server is optimized for high-throughput deployment using [vLLM](https://github.com/vllm-project/vllm) and can run on a consumer GPU with 24GB RAM.","release_date":"2024-05-07","context_window":8192,"license":"llama3","tagline":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below.","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"openchat/openchat-3.6-8b","name":"openchat-3.6-8b-20240522","created_by":"openchat","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["openchat/openchat-3.6-8b-20240522"],"parameters":8,"page_url":"https://huggingface.co/openchat/openchat-3.6-8b-20240522","description":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below. The server is optimized for high-throughput deployment using [vLLM](https://github.com/vllm-project/vllm) and can run on a consumer GPU with 24GB RAM.","release_date":"2024-05-07","context_window":8192,"license":"llama3","tagline":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below.","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"openchat/openchat_3.5","name":"openchat_3.5","created_by":"openchat","source":"official","description":"**🔥 The first 7B model Achieves Comparable Results with ChatGPT (March)! 🔥**","release_date":"2023-10-30","context_window":8192,"license":"apache-2.0","page_url":"https://huggingface.co/openchat/openchat_3.5","architecture":"transformer","tagline":"**🔥 The first 7B model Achieves Comparable Results with ChatGPT (March)! 🔥**","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"OpenPipe/Qwen3-14B-Instruct","name":"Qwen3-14B-Instruct","created_by":"openpipe","source":"official","release_date":"2025-10-10","context_window":40960,"license":"apache-2.0","parameters":15,"page_url":"https://huggingface.co/OpenPipe/Qwen3-14B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Orion-zhen/Qwen2.5-7B-Instruct-Uncensored","name":"Qwen2.5-7B-Instruct-Uncensored","created_by":"orion-zhen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Orion-zhen/Qwen2.5-7B-Instruct-Uncensored","description":"This model is an uncensored fine-tune version of Qwen2.5-7B-Instruct. However, I can still notice that though uncensored, the model fails to generate detailed descriptions on certain extreme scenarios, which might be associated with deletion on some pretrain datasets in Qwen's pretraining stage.","release_date":"2024-09-26","context_window":32768,"license":"gpl-3.0","tagline":"This model is an uncensored fine-tune version of Qwen2.5-7B-Instruct.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"PatronusAI/Llama-3-Patronus-Lynx-8B-Instruct-v1.1","name":"Llama-3-Patronus-Lynx-8B-Instruct-v1.1","created_by":"patronusai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/PatronusAI/Llama-3-Patronus-Lynx-8B-Instruct-v1.1","description":"Lynx is an open-source hallucination evaluation model. Patronus-Lynx-8B-Instruct-v1.1 was trained on a mix of datasets including CovidQA, PubmedQA, DROP, RAGTruth. The datasets contain a mix of hand-annotated and synthetic data. The maximum sequence length is 128000 tokens.","release_date":"2024-07-24","context_window":131072,"license":"cc-by-nc-4.0","tagline":"Lynx is an open-source hallucination evaluation model.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"prefeitura-rio/Rio-3.0-Open-Mini","name":"Rio-3.0-Open-Mini","created_by":"prefeitura-rio","source":"official","description":"**Rio 3.0 Open Mini** is a frontier-class reasoning model developed by [IplanRIO](https://iplanrio.rio.rj.gov.br/), the municipal IT company of Rio de Janeiro's city government. Built through distillation on top of Qwen3-4B-Thinking-2507 using reasoning traces from our to be announced Rio 3.0 model, Rio 3.0 Open achieves state-of-the-art results across mathematics, STEM, and code benchmarks — surpassing its base model by significant margins and competing with models far larger than itself.","release_date":"2026-02-09","context_window":262144,"license":"mit","parameters":4,"page_url":"https://huggingface.co/prefeitura-rio/Rio-3.0-Open-Mini","architecture":"transformer","tagline":"**Rio 3.0 Open Mini** is a frontier-class reasoning model developed by [IplanRIO](https://iplanrio.rio.rj.gov.br/), the municipal IT company of Rio de Janeiro's city government.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen1.5-0.5B-Chat","name":"Qwen1.5-0.5B-Chat","created_by":"qwen","source":"official","release_date":"2024-01-31","context_window":32768,"license":"other","parameters":0.62,"page_url":"https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen1.5-1.8B-Chat","name":"Qwen1.5-1.8B-Chat","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.8,"page_url":"https://huggingface.co/Qwen/Qwen1.5-1.8B-Chat","release_date":"2024-01-30","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2-0.5B-Instruct","name":"Qwen2-0.5B-Instruct","created_by":"qwen","source":"official","release_date":"2024-06-03","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2-0.5B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2-1.5B-Instruct","name":"Qwen2-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2-1.5B-Instruct","release_date":"2024-06-03","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2-72B-Instruct","name":"Qwen2-72B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":73,"page_url":"https://huggingface.co/Qwen/Qwen2-72B-Instruct","release_date":"2024-05-28","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","max_output_tokens":6144},{"id":"Qwen/Qwen2-7B-Instruct","name":"Qwen2-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2-7B-Instruct","release_date":"2024-06-04","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud."},{"id":"Qwen/Qwen2.5-0.5B-Instruct","name":"Qwen2.5-0.5B-Instruct","created_by":"qwen","source":"official","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-1.5B-Instruct","name":"Qwen2.5-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct","release_date":"2024-09-17","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-14B-Instruct-1M","name":"Qwen2.5-14B-Instruct-1M","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-14B-Instruct-1M","release_date":"2025-01-23","context_window":1010000,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","max_output_tokens":8192},{"id":"Qwen/Qwen2.5-14B-Instruct","name":"Qwen2.5-14B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-14B-Instruct","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-32B-Instruct","name":"Qwen2.5-32B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen2.5-32B-Instruct","release_date":"2024-09-17","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-3B-Instruct","name":"Qwen2.5-3B-Instruct","created_by":"qwen","source":"official","release_date":"2024-09-17","context_window":32768,"license":"other","parameters":3.1,"page_url":"https://huggingface.co/Qwen/Qwen2.5-3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-72B-Instruct","name":"Qwen2.5-72B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":73,"page_url":"https://huggingface.co/Qwen/Qwen2.5-72B-Instruct","release_date":"2024-09-16","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-7B-Instruct-1M","name":"Qwen2.5-7B-Instruct-1M","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-7B-Instruct-1M","release_date":"2025-01-23","context_window":1010000,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-7B-Instruct","name":"Qwen2.5-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-7B-Instruct","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-Coder-0.5B-Instruct","name":"Qwen2.5-Coder-0.5B-Instruct","created_by":"qwen","source":"official","release_date":"2024-11-06","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-0.5B-Instruct","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-Coder-0.5B","name":"Qwen2.5-Coder-0.5B","created_by":"qwen","source":"official","release_date":"2024-11-08","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-0.5B","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-Coder-1.5B-Instruct","name":"Qwen2.5-Coder-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-1.5B-Instruct","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-1.5B","name":"Qwen2.5-Coder-1.5B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-1.5B","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-14B-Instruct","name":"Qwen2.5-Coder-14B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-14B-Instruct","release_date":"2024-11-06","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-14B","name":"Qwen2.5-Coder-14B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-14B","release_date":"2024-11-08","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-32B-Instruct","name":"Qwen2.5-Coder-32B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct","release_date":"2024-11-06","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-32B","name":"Qwen2.5-Coder-32B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-32B","release_date":"2024-11-08","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-3B-Instruct","name":"Qwen2.5-Coder-3B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":3.1,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-3B-Instruct","release_date":"2024-11-06","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-3B","name":"Qwen2.5-Coder-3B","created_by":"qwen","source":"official","release_date":"2024-11-08","context_window":32768,"license":"other","parameters":3.1,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-3B","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-Coder-7B-Instruct","name":"Qwen2.5-Coder-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct","release_date":"2024-09-17","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-7B","name":"Qwen2.5-Coder-7B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-7B","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Math-1.5B-Instruct","name":"Qwen2.5-Math-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Math-1.5B-Instruct","description":"> [!Warning] >
](https://discord.gg/u54VK8m8tk) [](https://ko-fi.com/unsloth) [](https://github.com/unslothai/unsloth)","release_date":"2024-02-14","context_window":2048,"license":"apache-2.0","parameters":1.1,"tagline":"[](https://discord.gg/u54VK8m8tk) [](https://dicta.org.il)","release_date":"2025-12-01","context_window":62080,"license":"apache-2.0","page_url":"https://huggingface.co/dicta-il/DictaLM-3.0-1.7B-Instruct","architecture":"transformer","tagline":"[](https://dicta.org.il)","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"dphn/Dolphin-Mistral-24B-Venice-Edition","name":"Dolphin-Mistral-24B-Venice-Edition","created_by":"dphn","source":"official","description":"Website: https://dphn.ai Twitter: https://x.com/dphnAI Web Chat: https://chat.dphn.ai Telegram bot: https://t.me/DolphinAI_bot","release_date":"2025-06-12","context_window":32768,"license":"apache-2.0","parameters":24,"page_url":"https://huggingface.co/dphn/Dolphin-Mistral-24B-Venice-Edition","architecture":"transformer","tagline":"Website: https://dphn.ai Twitter: https://x.com/dphnAI Web Chat: https://chat.dphn.ai Telegram bot: https://t.me/DolphinAI_bot","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-07"},{"id":"Efficient-Large-Model/gemma-2-2b-it","name":"gemma-2-2b-it","created_by":"efficient-large-model","source":"official","last_updated":"2026-03-24","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":2.6,"page_url":"https://huggingface.co/Efficient-Large-Model/gemma-2-2b-it","description":"**Model Page**: [Gemma](https://ai.google.dev/gemma/docs/base)","release_date":"2024-12-12","context_window":8192,"license":"gemma","tagline":"**Model Page**: [Gemma](https://ai.google.dev/gemma/docs/base)","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"EssentialAI/rnj-1-instruct","name":"rnj-1-instruct","created_by":"essentialai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"page_url":"https://huggingface.co/EssentialAI/rnj-1-instruct","description":"Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models. These models perform well across a range of programming languages and boast strong agentic capabilities (e.g., inside agentic frameworks like mini-SWE-agent), while also excelling at tool-calling. They additionally exhibit strong capabilities in math and science.","release_date":"2025-12-04","context_window":32768,"license":"apache-2.0","parameters":8.3,"tagline":"style=\"vertical-align: middle;\" src=\"https://img.shields.io/badge/%F0%9F%8C%90%20Website-essential.ai-4b9fe1?color=4b9fe1&logoColor=white\"/>","architecture":"transformer","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"failspy/llama-3-70B-Instruct-abliterated","name":"llama-3-70B-Instruct-abliterated","created_by":"failspy","source":"official","family":"llama-3","description":"This is meta-llama/Llama-3-70B-Instruct with orthogonalized bfloat16 safetensor weights, generated with the methodology that was described in the preview paper/blog post: '[Refusal in LLMs is mediated by a single direction](https://www.alignmentforum.org/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction)' which I encourage you to read to understand more.","release_date":"2024-05-07","context_window":8192,"license":"llama3","page_url":"https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated","architecture":"transformer","model_type":"chat","tagline":"This is meta-llama/Llama-3-70B-Instruct with orthogonalized bfloat16 safetensor weights, generated with the methodology that was described in the preview paper/blog post: '[Refusal in LLMs is mediated","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-07"},{"id":"failspy/Meta-Llama-3-8B-Instruct-abliterated-v3","name":"Meta-Llama-3-8B-Instruct-abliterated-v3","created_by":"failspy","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/failspy/Meta-Llama-3-8B-Instruct-abliterated-v3","description":"[My Jupyter \"cookbook\" to replicate the methodology can be found here, refined library coming soon](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)","release_date":"2024-05-20","context_window":8192,"license":"llama3","tagline":"[My Jupyter \"cookbook\" to replicate the methodology can be found here, refined library coming soon](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"farbodtavakkoli/OTel-LLM-0.6B-IT","name":"OTel-LLM-0.6B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-0.6B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-0.6B-IT","architecture":"transformer","tagline":"**OTel-LLM-0.6B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-1.2B-IT","name":"OTel-LLM-1.2B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-1.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-03-10","context_window":128000,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-1.2B-IT","architecture":"transformer","tagline":"**OTel-LLM-1.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-1.7B-IT","name":"OTel-LLM-1.7B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-1.7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-1.7B-IT","architecture":"transformer","tagline":"**OTel-LLM-1.7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-12B-IT","name":"OTel-LLM-12B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-12B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-12B-IT","architecture":"transformer","tagline":"**OTel-LLM-12B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-22"},{"id":"farbodtavakkoli/OTel-LLM-14B-IT","name":"OTel-LLM-14B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-14B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-14B-IT","architecture":"transformer","tagline":"**OTel-LLM-14B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-1B-IT","name":"OTel-LLM-1B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-1B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":32768,"license":"apache-2.0","parameters":1,"page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-1B-IT","architecture":"transformer","tagline":"**OTel-LLM-1B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-20B-IT","name":"OTel-LLM-20B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-20B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-13","context_window":131072,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-20B-IT","architecture":"moe","tagline":"**OTel-LLM-20B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-270M-IT","name":"OTel-LLM-270M-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-270M-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":32768,"license":"apache-2.0","parameters":0.27,"page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-270M-IT","architecture":"transformer","tagline":"**OTel-LLM-270M-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-27B-IT","name":"OTel-LLM-27B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-27B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-27B-IT","architecture":"transformer","tagline":"**OTel-LLM-27B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-22"},{"id":"farbodtavakkoli/OTel-LLM-32B-IT","name":"OTel-LLM-32B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-32B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":65536,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-32B-IT","architecture":"transformer","tagline":"**OTel-LLM-32B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-4B-IT","name":"OTel-LLM-4B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-4B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","license":"apache-2.0","parameters":4.3,"page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-4B-IT","architecture":"transformer","tagline":"**OTel-LLM-4B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-7B-IT","name":"OTel-LLM-7B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":65536,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-7B-IT","architecture":"transformer","tagline":"**OTel-LLM-7B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-8.2B-IT","name":"OTel-LLM-8.2B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-8.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-8.2B-IT","architecture":"transformer","tagline":"**OTel-LLM-8.2B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"farbodtavakkoli/OTel-LLM-8.3B-IT","name":"OTel-LLM-8.3B-IT","created_by":"farbodtavakkoli","source":"official","description":"**OTel-LLM-8.3B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data. It is part of the [OTel Family of Models](https://huggingface.co/collections/farbodtavakkoli/otel-llm), an open-source initiative to build industry-standard AI models for the global telecommunications sector.","release_date":"2026-02-11","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/farbodtavakkoli/OTel-LLM-8.3B-IT","architecture":"transformer","tagline":"**OTel-LLM-8.3B-IT** is a telecom-specialized language model fine-tuned on telecommunications domain data.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"fdtn-ai/Foundation-Sec-8B-Instruct","name":"Foundation-Sec-8B-Instruct","created_by":"fdtn-ai","source":"official","release_date":"2025-07-29","context_window":131072,"license":"other","parameters":8,"page_url":"https://huggingface.co/fdtn-ai/Foundation-Sec-8B-Instruct","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"Featherless-Chat-Models/Mistral-7B-Instruct-v0.2","name":"Mistral-7B-Instruct-v0.2","created_by":"featherless-chat-models","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.2,"page_url":"https://huggingface.co/Featherless-Chat-Models/Mistral-7B-Instruct-v0.2","release_date":"2025-05-08","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"FlagAlpha/Llama3-Chinese-8B-Instruct","name":"Llama3-Chinese-8B-Instruct","created_by":"flagalpha","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/FlagAlpha/Llama3-Chinese-8B-Instruct","description":"Llama3-Chinese-8B-Instruct基于Llama3-8B中文微调对话模型，由Llama中文社区和AtomEcho（原子回声）联合研发，我们会持续提供更新的模型参数，模型训练过程见 [https://llama.family](https://llama.family)。","release_date":"2024-04-23","context_window":8192,"license":"apache-2.0","tagline":"Llama3-Chinese-8B-Instruct基于Llama3-8B中文微调对话模型，由Llama中文社区和AtomEcho（原子回声）联合研发，我们会持续提供更新的模型参数，模型训练过程见 [https://llama.family](https://llama.family)。","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-1.1-2b-it","name":"gemma-1.1-2b-it","created_by":"google","source":"official","family":"gemma-1","release_date":"2024-03-26","license":"gemma","parameters":2.5,"page_url":"https://huggingface.co/google/gemma-1.1-2b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"google/gemma-2-27b-it","name":"gemma-2-27b-it","created_by":"google","source":"official","family":"gemma-2","release_date":"2024-06-24","license":"gemma","parameters":27,"page_url":"https://huggingface.co/google/gemma-2-27b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"google/gemma-2-2b-it","name":"gemma-2-2b-it","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","release_date":"2024-07-16","license":"gemma","parameters":2.6,"page_url":"https://huggingface.co/google/gemma-2-2b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"google/gemma-2-2b-jpn-it","name":"gemma-2-2b-jpn-it","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":2.6,"page_url":"https://huggingface.co/google/gemma-2-2b-jpn-it","release_date":"2024-09-25","license":"gemma","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2-9b-it","name":"gemma-2-9b-it","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":9.2,"page_url":"https://huggingface.co/google/gemma-2-9b-it","release_date":"2024-06-24","license":"gemma","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2b-it","name":"gemma-2b-it","created_by":"google","source":"official","family":"gemma-2b","release_date":"2024-02-08","license":"gemma","page_url":"https://huggingface.co/google/gemma-2b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","parameters":2.5},{"id":"google/gemma-3-1b-it","name":"gemma-3-1b-it","created_by":"google","source":"official","family":"gemma-3","release_date":"2025-03-10","license":"gemma","parameters":1,"page_url":"https://huggingface.co/google/gemma-3-1b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"google/gemma-7b-it","name":"gemma-7b-it","created_by":"google","source":"official","family":"gemma-7b","release_date":"2024-02-13","license":"gemma","page_url":"https://huggingface.co/google/gemma-7b-it","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18","parameters":8.5},{"id":"google/medgemma-27b-text-it","name":"medgemma-27b-text-it","created_by":"google","source":"official","release_date":"2025-05-19","license":"other","page_url":"https://huggingface.co/google/medgemma-27b-text-it","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-29","parameters":27},{"id":"GraySwanAI/Llama-3-8B-Instruct-RR","name":"Llama-3-8B-Instruct-RR","created_by":"grayswanai","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/GraySwanAI/Llama-3-8B-Instruct-RR","description":"Llama-3-8B-Instruct-RR is a Llama-3 model with circuit breakers inserted using Representation Rerouting (RR).","release_date":"2024-07-08","context_window":8192,"tagline":"Llama-3-8B-Instruct-RR is a Llama-3 model with circuit breakers inserted using Representation Rerouting (RR).","architecture":"transformer","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"GritLM/GritLM-7B","name":"GritLM-7B","created_by":"gritlm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":7.2,"page_url":"https://huggingface.co/GritLM/GritLM-7B","description":"> GritLM is a generative representational instruction tuned language model. It unifies text representation (embedding) and text generation into a single model achieving state-of-the-art performance on both types of tasks.","release_date":"2024-02-11","context_window":32768,"license":"apache-2.0","tagline":"> GritLM is a generative representational instruction tuned language model.","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"h2oai/h2ogpt-4096-llama2-7b-chat","name":"h2ogpt-4096-llama2-7b-chat","created_by":"h2oai","source":"official","description":"h2oGPT clone of [Meta's Llama 2 7B Chat](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf).","release_date":"2023-08-09","context_window":4096,"license":"llama2","parameters":6.7,"page_url":"https://huggingface.co/h2oai/h2ogpt-4096-llama2-7b-chat","architecture":"transformer","tagline":"h2oGPT clone of [Meta's Llama 2 7B Chat](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf).","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"last_updated":"2026-05-17"},{"id":"huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated","name":"DeepSeek-R1-Distill-Qwen-32B-abliterated","created_by":"huihui-ai","source":"official","description":"This is an uncensored version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.","release_date":"2025-01-22","context_window":131072,"parameters":33,"page_url":"https://huggingface.co/huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated","architecture":"transformer","model_type":"reasoning","tagline":"This is an uncensored version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it).","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-14"},{"id":"huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2","name":"Qwen2.5-14B-Instruct-abliterated-v2","created_by":"huihui-ai","source":"official","description":"This is an uncensored version of [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","release_date":"2024-10-09","context_window":32768,"license":"apache-2.0","parameters":15,"page_url":"https://huggingface.co/huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2","architecture":"transformer","model_type":"chat","tagline":"This is an uncensored version of [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteratio","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-21"},{"id":"huihui-ai/Qwen2.5-14B-Instruct-abliterated","name":"Qwen2.5-14B-Instruct-abliterated","created_by":"huihui-ai","source":"official","description":"This is an uncensored version of [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","release_date":"2024-09-25","context_window":32768,"license":"apache-2.0","parameters":15,"page_url":"https://huggingface.co/huihui-ai/Qwen2.5-14B-Instruct-abliterated","architecture":"transformer","model_type":"chat","tagline":"This is an uncensored version of [Qwen/Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/ablite","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"huihui-ai/Qwen2.5-72B-Instruct-abliterated","name":"Qwen2.5-72B-Instruct-abliterated","created_by":"huihui-ai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":73,"page_url":"https://huggingface.co/huihui-ai/Qwen2.5-72B-Instruct-abliterated","description":"This is an uncensored version of [Qwen/Qwen2.5-72B-Instruct](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.","release_date":"2024-10-26","context_window":32768,"license":"other","tagline":"This is an uncensored version of [Qwen/Qwen2.5-72B-Instruct](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"instruction-pretrain/finance-Llama3-8B","name":"finance-Llama3-8B","created_by":"instruction-pretrain","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"parameters":8,"page_url":"https://huggingface.co/instruction-pretrain/finance-Llama3-8B","description":"This repo contains the **finance model developed from Llama3-8B** in our paper [Instruction Pre-Training: Language Models are Supervised Multitask Learners](https://huggingface.co/papers/2406.14491).","release_date":"2024-06-18","context_window":8192,"license":"llama3","tagline":"This repo contains the **finance model developed from Llama3-8B** in our paper [Instruction Pre-Training: Language Models are Supervised Multitask Learners](https://huggingface.co/papers/2406.14491).","architecture":"transformer","open_weight":true},{"id":"Intel/neural-chat-7b-v3-3","name":"neural-chat-7b-v3-3","created_by":"intel","source":"official","last_updated":"2026-03-22","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":7.2,"page_url":"https://huggingface.co/Intel/neural-chat-7b-v3-3","release_date":"2023-12-09","context_window":32768,"license":"apache-2.0"},{"id":"invalid-coder/Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp","name":"Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp","created_by":"invalid-coder","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":11,"page_url":"https://huggingface.co/invalid-coder/Sakura-SOLAR-Instruct-CarbonVillain-en-10.7B-v2-slerp","description":"NeuralPipe-7B-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [jeonsworld/CarbonVillain-en-10.7B-v2](https://huggingface.co/jeonsworld/CarbonVillain-en-10.7B-v2) * [kyujinpy/Sakura-SOLAR-Instruct](https://huggingface.co/kyujinpy/Sakura-SOLAR-Instruct)","release_date":"2024-01-10","context_window":4096,"license":"apache-2.0","tagline":"NeuralPipe-7B-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [jeonsworld/CarbonVillain-en-10.7B","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"kakaocorp/kanana-1.5-8b-instruct-2505","name":"kanana-1.5-8b-instruct-2505","created_by":"kakaocorp","source":"official","release_date":"2025-05-21","context_window":32768,"license":"apache-2.0","alias":"kakaocorp/kanana-1.5-8b-instruct","page_url":"https://huggingface.co/kakaocorp/kanana-1.5-8b-instruct-2505","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"kakaocorp/kanana-1.5-8b-instruct","name":"kanana-1.5-8b-instruct-2505","created_by":"kakaocorp","source":"official","release_date":"2025-05-21","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/kakaocorp/kanana-1.5-8b-instruct-2505","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"snapshots":["kakaocorp/kanana-1.5-8b-instruct-2505"],"last_updated":"2026-06-10"},{"id":"llmfan46/Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic","name":"Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic","created_by":"llmfan46","source":"official","release_date":"2026-04-25","license":"apache-2.0","parameters":31,"page_url":"https://huggingface.co/llmfan46/Gemma-4-Garnet-V2-31B-it-ultra-uncensored-heretic","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"lmstudio-community/Qwen3-1.7B-MLX-bf16","name":"Qwen3-1.7B-MLX-bf16","created_by":"lmstudio-community","source":"official","description":"This model [lmstudio-community/Qwen3-1.7B-MLX-bf16](https://huggingface.co/lmstudio-community/Qwen3-1.7B-MLX-bf16) was converted to MLX format from [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) using mlx-lm version **0.24.0**.","release_date":"2025-04-28","context_window":40960,"license":"apache-2.0","parameters":1.7,"page_url":"https://huggingface.co/lmstudio-community/Qwen3-1.7B-MLX-bf16","architecture":"transformer","model_type":"chat","tagline":"This model [lmstudio-community/Qwen3-1.7B-MLX-bf16](https://huggingface.co/lmstudio-community/Qwen3-1.7B-MLX-bf16) was converted to MLX format from [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"m8than/Mistral-Nemo-Instruct-2407-lenient-chatfix","name":"Mistral-Nemo-Instruct-2407-lenient-chatfix","created_by":"m8than","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","page_url":"https://huggingface.co/m8than/Mistral-Nemo-Instruct-2407-lenient-chatfix","description":"It's just mistral nemo 2407 with a less strict chat format.","release_date":"2025-05-06","context_window":131072,"license":"apache-2.0","parameters":12,"tagline":"It's just mistral nemo 2407 with a less strict chat format.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta-llama/Llama-2-13b-chat-hf","name":"Llama-2-13b-chat-hf","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-2","model_type":"chat","parameters":13,"page_url":"https://huggingface.co/meta-llama/Llama-2-13b-chat-hf","release_date":"2023-07-13","license":"llama2","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":4096,"max_output_tokens":4096},{"id":"meta-llama/Llama-3.1-70B-Instruct","name":"Llama-3.1-70B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct","release_date":"2024-07-16","license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.1-8B-Instruct","name":"Llama-3.1-8B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct","release_date":"2024-07-18","license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.2-1B-Instruct","name":"Llama-3.2-1B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.2","model_type":"chat","parameters":1.2,"page_url":"https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct","release_date":"2024-09-18","license":"llama3.2","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.2-3B-Instruct","name":"Llama-3.2-3B-Instruct","created_by":"meta","source":"official","family":"Llama-3.2","release_date":"2024-09-18","license":"llama3.2","parameters":3.2,"page_url":"https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-14","context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Llama-3.3-70B-Instruct","name":"Llama-3.3-70B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.3","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct","release_date":"2024-11-26","license":"llama3.3","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12"},{"id":"meta-llama/Meta-Llama-3-70B-Instruct","name":"Meta-Llama-3-70B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct","release_date":"2024-04-17","license":"llama3","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":8192,"max_output_tokens":8192},{"id":"meta-llama/Meta-Llama-3-8B-Instruct","name":"Meta-Llama-3-8B-Instruct","created_by":"meta","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct","release_date":"2024-04-17","license":"llama3","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":8192,"max_output_tokens":8192},{"id":"microsoft/Phi-3-mini-128k-instruct","name":"Phi-3-mini-128k-instruct","created_by":"microsoft","source":"official","description":"🎉**Phi-4**: [[multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx)]; [[mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-mini-instruct-onnx)]","release_date":"2024-04-22","context_window":131072,"license":"mit","parameters":3.8,"page_url":"https://huggingface.co/microsoft/Phi-3-mini-128k-instruct","architecture":"transformer","model_type":"chat","tagline":"🎉**Phi-4**: [[multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx)]; [[mini-instruct](https://hug","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"microsoft/Phi-3-mini-4k-instruct","name":"Phi-3-mini-4k-instruct","created_by":"microsoft","source":"official","description":"🎉 **Phi-3.5**: [[mini-instruct]](https://huggingface.co/microsoft/Phi-3.5-mini-instruct); [[MoE-instruct]](https://huggingface.co/microsoft/Phi-3.5-MoE-instruct) ; [[vision-instruct]](https://huggingface.co/microsoft/Phi-3.5-vision-instruct)","release_date":"2024-04-22","context_window":4096,"license":"mit","parameters":3.8,"page_url":"https://huggingface.co/microsoft/Phi-3-mini-4k-instruct","architecture":"transformer","model_type":"chat","tagline":"🎉 **Phi-3.5**: [[mini-instruct]](https://huggingface.co/microsoft/Phi-3.5-mini-instruct); [[MoE-instruct]](https://huggingface.co/microsoft/Phi-3.5-MoE-instruct) ; [[vision-instruct]](https://hugging","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"microsoft/Phi-4-mini-instruct","name":"Phi-4-mini-instruct","created_by":"microsoft","source":"official","description":"🎉**Phi-4**: [[mini-reasoning](https://huggingface.co/microsoft/Phi-4-mini-reasoning) | [reasoning](https://huggingface.co/microsoft/Phi-4-reasoning)] | [[multimodal-instruct](https://huggingface.co/microsoft/Phi-4-multimodal-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-multimodal-instruct-onnx)]; [[mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) | [onnx](https://huggingface.co/microsoft/Phi-4-mini-instruct-onnx)]","release_date":"2025-02-19","context_window":131072,"license":"mit","page_url":"https://huggingface.co/microsoft/Phi-4-mini-instruct","architecture":"transformer","model_type":"chat","tagline":"🎉**Phi-4**: [[mini-reasoning](https://huggingface.co/microsoft/Phi-4-mini-reasoning) | [reasoning](https://huggingface.co/microsoft/Phi-4-reasoning)] | [[multimodal-instruct](https://huggingface.co/m","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","parameters":3.8},{"id":"mistralai/Mistral-7B-Instruct-v0.1","name":"Mistral-7B-Instruct-v0.1","created_by":"mistral","source":"official","release_date":"2023-09-27","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","parameters":7.2},{"id":"mistralai/Mistral-7B-Instruct-v0.2","name":"Mistral-7B-Instruct-v0.2","created_by":"mistral","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.2,"page_url":"https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2","release_date":"2023-12-11","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated","name":"Meta-Llama-3.1-8B-Instruct-abliterated","created_by":"mlabonne","source":"official","last_updated":"2026-03-24","family":"Llama-3.1","page_url":"https://huggingface.co/mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"This is an uncensored version of Llama 3.1 8B Instruct created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","release_date":"2024-07-24","context_window":131072,"license":"llama3.1","parameters":8,"tagline":"This is an uncensored version of Llama 3.1 8B Instruct created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mlabonne/NeuralDaredevil-8B-abliterated","name":"NeuralDaredevil-8B-abliterated","created_by":"mlabonne","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":8,"page_url":"https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated","description":"This is a DPO fine-tune of [mlabonne/Daredevil-8-abliterated](https://huggingface.co/mlabonne/Daredevil-8B-abliterated), trained on one epoch of [mlabonne/orpo-dpo-mix-40k](https://huggingface.co/datasets/mlabonne/orpo-dpo-mix-40k). The DPO fine-tuning successfully recovers the performance loss due to the abliteration process, making it an excellent uncensored model.","release_date":"2024-05-27","context_window":8192,"license":"llama3","tagline":"This is a DPO fine-tune of [mlabonne/Daredevil-8-abliterated](https://huggingface.co/mlabonne/Daredevil-8B-abliterated), trained on one epoch of [mlabonne/orpo-dpo-mix-40k](https://huggingface.co/datasets/mlabonne/orpo-dpo-mix-40k).","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"moonshotai/Kimi-K2-Instruct-0905","name":"Kimi-K2-Instruct-0905","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Kimi-K2","model_type":"chat","page_url":"https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905","release_date":"2025-09-03","context_window":262144,"license":"other","parameters":1026,"architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/Kimi-K2-Instruct","name":"Kimi-K2-Instruct","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Kimi-K2","model_type":"chat","page_url":"https://huggingface.co/moonshotai/Kimi-K2-Instruct","release_date":"2025-07-11","context_window":131072,"license":"other","parameters":1026,"architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/Kimi-K2-Thinking","name":"Kimi-K2-Thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true},"family":"Kimi-K2","model_type":"chat","page_url":"https://huggingface.co/moonshotai/Kimi-K2-Thinking","release_date":"2025-11-04","context_window":262144,"license":"other","parameters":1058,"architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Long-term thinking, multi-step tool usage, complex problem solving.","tagline":"Long-term thinking, multi-step tool usage, complex problem solving.","max_output_tokens":32768},{"id":"moonshotai/Kimi-Linear-48B-A3B-Instruct","name":"Kimi-Linear-48B-A3B-Instruct","created_by":"moonshot","source":"official","family":"kimi","release_date":"2025-10-30","license":"mit","parameters":49,"page_url":"https://huggingface.co/moonshotai/Kimi-Linear-48B-A3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"NousResearch/Meta-Llama-3-70B-Instruct","name":"Meta-Llama-3-70B-Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":71,"page_url":"https://huggingface.co/NousResearch/Meta-Llama-3-70B-Instruct","release_date":"2024-04-19","context_window":8192,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"NousResearch/Meta-Llama-3-8B-Instruct","name":"Meta-Llama-3-8B-Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct","release_date":"2024-04-18","context_window":8192,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"NousResearch/Meta-Llama-3.1-8B-Instruct","name":"Meta-Llama-3.1-8B-Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/NousResearch/Meta-Llama-3.1-8B-Instruct","release_date":"2024-07-24","context_window":131072,"license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/Gemma-4-31B-IT-NVFP4","name":"Gemma-4-31B-IT-NVFP4","created_by":"nvidia","source":"official","release_date":"2026-04-02","license":"other","parameters":21,"page_url":"https://huggingface.co/nvidia/Gemma-4-31B-IT-NVFP4","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"nvidia/Llama-3.1-8B-Instruct-FP8","name":"Llama-3.1-8B-Instruct-FP8","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/nvidia/Llama-3.1-8B-Instruct-FP8","release_date":"2024-08-29","context_window":131072,"license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"structured_output":true,"reasoning":true,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":120,"page_url":"https://huggingface.co/openai/gpt-oss-120b","description":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","release_date":"2025-08-04","context_window":131072,"license":"apache-2.0","tagline":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":131072,"knowledge_cutoff":"2024-06"},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"structured_output":true,"reasoning":true,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":22,"page_url":"https://huggingface.co/openai/gpt-oss-20b","description":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","release_date":"2025-08-04","context_window":131072,"license":"apache-2.0","tagline":"Welcome to the gpt-oss series, [OpenAI’s open-weight models](https://openai.com/open-models) designed for powerful reasoning, agentic tasks, and versatile developer use cases.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":131072,"knowledge_cutoff":"2024-06"},{"id":"openai/gpt-oss-safeguard-20b","name":"gpt-oss-safeguard-20b","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"moderation","parameters":22,"page_url":"https://huggingface.co/openai/gpt-oss-safeguard-20b","description":"`gpt-oss-safeguard-120b` and `gpt-oss-safeguard-20b` are safety reasoning models built-upon gpt-oss. With these models, you can classify text content based on safety policies that you provide and perform a suite of foundational safety tasks. These models are intended for safety use cases. For other applications, we recommend using [gpt-oss models](https://huggingface.co/collections/openai/gpt-oss).","release_date":"2025-09-18","context_window":131072,"license":"apache-2.0","tagline":"`gpt-oss-safeguard-120b` and `gpt-oss-safeguard-20b` are safety reasoning models built-upon gpt-oss.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openchat/openchat-3.6-8b-20240522","name":"openchat-3.6-8b-20240522","created_by":"openchat","source":"official","last_updated":"2026-03-24","alias":"openchat/openchat-3.6-8b","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":8,"page_url":"https://huggingface.co/openchat/openchat-3.6-8b-20240522","description":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below. The server is optimized for high-throughput deployment using [vLLM](https://github.com/vllm-project/vllm) and can run on a consumer GPU with 24GB RAM.","release_date":"2024-05-07","context_window":8192,"license":"llama3","tagline":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below.","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"openchat/openchat-3.6-8b","name":"openchat-3.6-8b-20240522","created_by":"openchat","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["openchat/openchat-3.6-8b-20240522"],"parameters":8,"page_url":"https://huggingface.co/openchat/openchat-3.6-8b-20240522","description":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below. The server is optimized for high-throughput deployment using [vLLM](https://github.com/vllm-project/vllm) and can run on a consumer GPU with 24GB RAM.","release_date":"2024-05-07","context_window":8192,"license":"llama3","tagline":"To use this model, we highly recommend installing the OpenChat package by following the [installation guide](https://github.com/imoneoi/openchat#installation) in our repository and using the OpenChat OpenAI-compatible API server by running the serving command from the table below.","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"openchat/openchat_3.5","name":"openchat_3.5","created_by":"openchat","source":"official","description":"**🔥 The first 7B model Achieves Comparable Results with ChatGPT (March)! 🔥**","release_date":"2023-10-30","context_window":8192,"license":"apache-2.0","page_url":"https://huggingface.co/openchat/openchat_3.5","architecture":"transformer","tagline":"**🔥 The first 7B model Achieves Comparable Results with ChatGPT (March)! 🔥**","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"OpenPipe/Qwen3-14B-Instruct","name":"Qwen3-14B-Instruct","created_by":"openpipe","source":"official","release_date":"2025-10-10","context_window":40960,"license":"apache-2.0","parameters":15,"page_url":"https://huggingface.co/OpenPipe/Qwen3-14B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Orion-zhen/Qwen2.5-7B-Instruct-Uncensored","name":"Qwen2.5-7B-Instruct-Uncensored","created_by":"orion-zhen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Orion-zhen/Qwen2.5-7B-Instruct-Uncensored","description":"This model is an uncensored fine-tune version of Qwen2.5-7B-Instruct. However, I can still notice that though uncensored, the model fails to generate detailed descriptions on certain extreme scenarios, which might be associated with deletion on some pretrain datasets in Qwen's pretraining stage.","release_date":"2024-09-26","context_window":32768,"license":"gpl-3.0","tagline":"This model is an uncensored fine-tune version of Qwen2.5-7B-Instruct.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"PatronusAI/Llama-3-Patronus-Lynx-8B-Instruct-v1.1","name":"Llama-3-Patronus-Lynx-8B-Instruct-v1.1","created_by":"patronusai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/PatronusAI/Llama-3-Patronus-Lynx-8B-Instruct-v1.1","description":"Lynx is an open-source hallucination evaluation model. Patronus-Lynx-8B-Instruct-v1.1 was trained on a mix of datasets including CovidQA, PubmedQA, DROP, RAGTruth. The datasets contain a mix of hand-annotated and synthetic data. The maximum sequence length is 128000 tokens.","release_date":"2024-07-24","context_window":131072,"license":"cc-by-nc-4.0","tagline":"Lynx is an open-source hallucination evaluation model.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"prefeitura-rio/Rio-3.0-Open-Mini","name":"Rio-3.0-Open-Mini","created_by":"prefeitura-rio","source":"official","description":"**Rio 3.0 Open Mini** is a frontier-class reasoning model developed by [IplanRIO](https://iplanrio.rio.rj.gov.br/), the municipal IT company of Rio de Janeiro's city government. Built through distillation on top of Qwen3-4B-Thinking-2507 using reasoning traces from our to be announced Rio 3.0 model, Rio 3.0 Open achieves state-of-the-art results across mathematics, STEM, and code benchmarks — surpassing its base model by significant margins and competing with models far larger than itself.","release_date":"2026-02-09","context_window":262144,"license":"mit","parameters":4,"page_url":"https://huggingface.co/prefeitura-rio/Rio-3.0-Open-Mini","architecture":"transformer","tagline":"**Rio 3.0 Open Mini** is a frontier-class reasoning model developed by [IplanRIO](https://iplanrio.rio.rj.gov.br/), the municipal IT company of Rio de Janeiro's city government.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen1.5-0.5B-Chat","name":"Qwen1.5-0.5B-Chat","created_by":"qwen","source":"official","release_date":"2024-01-31","context_window":32768,"license":"other","parameters":0.62,"page_url":"https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen1.5-1.8B-Chat","name":"Qwen1.5-1.8B-Chat","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.8,"page_url":"https://huggingface.co/Qwen/Qwen1.5-1.8B-Chat","release_date":"2024-01-30","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2-0.5B-Instruct","name":"Qwen2-0.5B-Instruct","created_by":"qwen","source":"official","release_date":"2024-06-03","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2-0.5B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2-1.5B-Instruct","name":"Qwen2-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2-1.5B-Instruct","release_date":"2024-06-03","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2-72B-Instruct","name":"Qwen2-72B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":73,"page_url":"https://huggingface.co/Qwen/Qwen2-72B-Instruct","release_date":"2024-05-28","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","max_output_tokens":6144},{"id":"Qwen/Qwen2-7B-Instruct","name":"Qwen2-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2-7B-Instruct","release_date":"2024-06-04","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud."},{"id":"Qwen/Qwen2.5-0.5B-Instruct","name":"Qwen2.5-0.5B-Instruct","created_by":"qwen","source":"official","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-1.5B-Instruct","name":"Qwen2.5-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct","release_date":"2024-09-17","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-14B-Instruct-1M","name":"Qwen2.5-14B-Instruct-1M","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-14B-Instruct-1M","release_date":"2025-01-23","context_window":1010000,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","max_output_tokens":8192},{"id":"Qwen/Qwen2.5-14B-Instruct","name":"Qwen2.5-14B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-14B-Instruct","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-32B-Instruct","name":"Qwen2.5-32B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen2.5-32B-Instruct","release_date":"2024-09-17","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-3B-Instruct","name":"Qwen2.5-3B-Instruct","created_by":"qwen","source":"official","release_date":"2024-09-17","context_window":32768,"license":"other","parameters":3.1,"page_url":"https://huggingface.co/Qwen/Qwen2.5-3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-72B-Instruct","name":"Qwen2.5-72B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":73,"page_url":"https://huggingface.co/Qwen/Qwen2.5-72B-Instruct","release_date":"2024-09-16","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-7B-Instruct-1M","name":"Qwen2.5-7B-Instruct-1M","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-7B-Instruct-1M","release_date":"2025-01-23","context_window":1010000,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-7B-Instruct","name":"Qwen2.5-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-7B-Instruct","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen."},{"id":"Qwen/Qwen2.5-Coder-0.5B-Instruct","name":"Qwen2.5-Coder-0.5B-Instruct","created_by":"qwen","source":"official","release_date":"2024-11-06","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-0.5B-Instruct","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-Coder-0.5B","name":"Qwen2.5-Coder-0.5B","created_by":"qwen","source":"official","release_date":"2024-11-08","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-0.5B","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-Coder-1.5B-Instruct","name":"Qwen2.5-Coder-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-1.5B-Instruct","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-1.5B","name":"Qwen2.5-Coder-1.5B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-1.5B","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-14B-Instruct","name":"Qwen2.5-Coder-14B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-14B-Instruct","release_date":"2024-11-06","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-14B","name":"Qwen2.5-Coder-14B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-14B","release_date":"2024-11-08","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-32B-Instruct","name":"Qwen2.5-Coder-32B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct","release_date":"2024-11-06","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-32B","name":"Qwen2.5-Coder-32B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-32B","release_date":"2024-11-08","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-3B-Instruct","name":"Qwen2.5-Coder-3B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":3.1,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-3B-Instruct","release_date":"2024-11-06","context_window":32768,"license":"other","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-3B","name":"Qwen2.5-Coder-3B","created_by":"qwen","source":"official","release_date":"2024-11-08","context_window":32768,"license":"other","parameters":3.1,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-3B","architecture":"transformer","model_type":"code","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen2.5-Coder-7B-Instruct","name":"Qwen2.5-Coder-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct","release_date":"2024-09-17","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Coder-7B","name":"Qwen2.5-Coder-7B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Coder-7B","release_date":"2024-09-16","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Math-1.5B-Instruct","name":"Qwen2.5-Math-1.5B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1.5,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Math-1.5B-Instruct","description":"> [!Warning] > > > 🚨 Qwen2.5-Math mainly supports solving English and Chinese math problems through CoT and TIR. We do not recommend using this series of models for other tasks. > >
","release_date":"2024-09-16","context_window":4096,"license":"apache-2.0","tagline":"> [!Warning] > > > 🚨 Qwen2.5-Math mainly supports solving English and Chinese math problems through CoT and TIR.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen2.5-Math-7B-Instruct","name":"Qwen2.5-Math-7B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/Qwen/Qwen2.5-Math-7B-Instruct","description":"> [!Warning] >
> > 🚨 Qwen2.5-Math mainly supports solving English and Chinese math problems through CoT and TIR. We do not recommend using this series of models for other tasks. > >
","release_date":"2024-09-19","context_window":4096,"license":"apache-2.0","tagline":"> [!Warning] > > > 🚨 Qwen2.5-Math mainly supports solving English and Chinese math problems through CoT and TIR.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-0.6B-Base","name":"Qwen3-0.6B-Base","created_by":"qwen","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/Qwen/Qwen3-0.6B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","parameters":0.6},{"id":"Qwen/Qwen3-0.6B","name":"Qwen3-0.6B","created_by":"qwen","source":"official","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","parameters":0.75,"page_url":"https://huggingface.co/Qwen/Qwen3-0.6B","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-1.7B-Base","name":"Qwen3-1.7B-Base","created_by":"qwen","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","parameters":1.7,"page_url":"https://huggingface.co/Qwen/Qwen3-1.7B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-1.7B","name":"Qwen3-1.7B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":2,"page_url":"https://huggingface.co/Qwen/Qwen3-1.7B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-14B-Base","name":"Qwen3-14B-Base","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-14B-Base","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-14B","name":"Qwen3-14B","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-14B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Instruct-2507","name":"Qwen3-235B-A22B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","alias":"Qwen/Qwen3-235B-A22B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507","release_date":"2025-07-21","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Instruct","name":"Qwen3-235B-A22B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"snapshots":["Qwen/Qwen3-235B-A22B-Instruct-2507"],"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507","release_date":"2025-07-21","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Thinking-2507","name":"Qwen3-235B-A22B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","alias":"Qwen/Qwen3-235B-A22B-Thinking","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507","release_date":"2025-07-25","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B-Thinking","name":"Qwen3-235B-A22B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"snapshots":["Qwen/Qwen3-235B-A22B-Thinking-2507"],"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507","release_date":"2025-07-25","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-235B-A22B","name":"Qwen3-235B-A22B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":235,"active_parameters":15,"page_url":"https://huggingface.co/Qwen/Qwen3-235B-A22B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-30B-A3B-Instruct-2507","name":"Qwen3-30B-A3B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","alias":"Qwen/Qwen3-30B-A3B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507","release_date":"2025-07-28","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-30B-A3B-Instruct","name":"Qwen3-30B-A3B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"snapshots":["Qwen/Qwen3-30B-A3B-Instruct-2507"],"model_type":"chat","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507","release_date":"2025-07-28","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-30B-A3B-Thinking-2507","name":"Qwen3-30B-A3B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","release_date":"2025-07-29","context_window":262144,"model_type":"chat","license":"apache-2.0","parameters":31,"active_parameters":1.9,"alias":"Qwen/Qwen3-30B-A3B-Thinking","page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507","architecture":"moe","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"tools":["function_calling"]},{"id":"Qwen/Qwen3-30B-A3B-Thinking","name":"Qwen3-30B-A3B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","release_date":"2025-07-29","context_window":262144,"license":"apache-2.0","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"snapshots":["Qwen/Qwen3-30B-A3B-Thinking-2507"]},{"id":"Qwen/Qwen3-30B-A3B","name":"Qwen3-30B-A3B","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-30B-A3B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-32B","name":"Qwen3-32B","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":33,"page_url":"https://huggingface.co/Qwen/Qwen3-32B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-Base","name":"Qwen3-4B-Base","created_by":"qwen","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-4B-Instruct-2507","name":"Qwen3-4B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","alias":"Qwen/Qwen3-4B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-Instruct","name":"Qwen3-4B-Instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["Qwen/Qwen3-4B-Instruct-2507"],"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-SafeRL","name":"Qwen3-4B-SafeRL","created_by":"qwen","source":"official","release_date":"2025-09-30","context_window":40960,"license":"apache-2.0","parameters":4.4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-SafeRL","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3-4B-Thinking-2507","name":"Qwen3-4B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","alias":"Qwen/Qwen3-4B-Thinking","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B-Thinking","name":"Qwen3-4B-Thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["Qwen/Qwen3-4B-Thinking-2507"],"model_type":"chat","parameters":4,"page_url":"https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507","release_date":"2025-08-05","context_window":262144,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-4B","name":"Qwen3-4B","created_by":"qwen","source":"official","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","page_url":"https://huggingface.co/Qwen/Qwen3-4B","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","parameters":4},{"id":"Qwen/Qwen3-8B-Base","name":"Qwen3-8B-Base","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8.2,"page_url":"https://huggingface.co/Qwen/Qwen3-8B-Base","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-8B","name":"Qwen3-8B","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":8.2,"page_url":"https://huggingface.co/Qwen/Qwen3-8B","release_date":"2025-04-27","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-30B-A3B-Instruct","name":"Qwen3-Coder-30B-A3B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","parameters":31,"active_parameters":1.9,"page_url":"https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct","release_date":"2025-07-31","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8","name":"Qwen3-Coder-480B-A35B-Instruct-FP8","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":480,"active_parameters":24,"release_date":"2025-07-22","context_window":262144,"license":"apache-2.0","page_url":"https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-480B-A35B-Instruct","name":"Qwen3-Coder-480B-A35B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","parameters":480,"active_parameters":24,"page_url":"https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct","release_date":"2025-07-22","context_window":262144,"license":"apache-2.0","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-Next-FP8","name":"Qwen3-Coder-Next-FP8","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","page_url":"https://huggingface.co/Qwen/Qwen3-Coder-Next-FP8","release_date":"2026-02-01","context_window":262144,"license":"apache-2.0","parameters":80,"active_parameters":1.6,"architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Coder-Next","name":"Qwen3-Coder-Next","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","release_date":"2026-01-30","context_window":262144,"license":"apache-2.0","parameters":80,"active_parameters":1.6,"page_url":"https://huggingface.co/Qwen/Qwen3-Coder-Next","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","max_output_tokens":65536},{"id":"Qwen/Qwen3-Next-80B-A3B-Instruct","name":"Qwen3-Next-80B-A3B-Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":81,"active_parameters":1.6,"page_url":"https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct","description":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI). We are excited to share our latest advancements in addressing these demands, centered on improving scaling efficiency through innovative model architecture. We call this next-generation foundation models **Qwen3-Next**.","release_date":"2025-09-09","context_window":262144,"license":"apache-2.0","tagline":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI).","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3-Next-80B-A3B-Thinking","name":"Qwen3-Next-80B-A3B-Thinking","created_by":"qwen","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":81,"active_parameters":1.6,"page_url":"https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking","description":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI). We are excited to share our latest advancements in addressing these demands, centered on improving scaling efficiency through innovative model architecture. We call this next-generation foundation models **Qwen3-Next**.","release_date":"2025-09-09","context_window":262144,"license":"apache-2.0","tagline":"Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI).","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"Qwen/Qwen3Guard-Gen-0.6B","name":"Qwen3Guard-Gen-0.6B","created_by":"qwen","source":"official","description":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: **Qwen3Guard-Gen**, a generative model that frames safety classification as an instruction-following task, and **Qwen3Guard-Stream**, which incorporates a token-level classification head for real-time safety monitoring during incremental text ge…","release_date":"2025-09-23","context_window":32768,"license":"apache-2.0","parameters":0.75,"page_url":"https://huggingface.co/Qwen/Qwen3Guard-Gen-0.6B","architecture":"transformer","model_type":"chat","tagline":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"Qwen/Qwen3Guard-Gen-4B","name":"Qwen3Guard-Gen-4B","created_by":"qwen","source":"official","description":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: **Qwen3Guard-Gen**, a generative model that frames safety classification as an instruction-following task, and **Qwen3Guard-Stream**, which incorporates a token-level classification head for real-time safety monitoring during incremental text ge…","release_date":"2025-09-23","context_window":32768,"license":"apache-2.0","parameters":4.4,"page_url":"https://huggingface.co/Qwen/Qwen3Guard-Gen-4B","architecture":"transformer","model_type":"chat","tagline":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"Qwen/Qwen3Guard-Gen-8B","name":"Qwen3Guard-Gen-8B","created_by":"qwen","source":"official","description":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety. The series includes models of three sizes (0.6B, 4B, and 8B) and features two specialized variants: **Qwen3Guard-Gen**, a generative model that frames safety classification as an instruction-following task, and **Qwen3Guard-Stream**, which incorporates a token-level classification head for real-time safety monitoring during incremental text ge…","release_date":"2025-09-23","context_window":32768,"license":"apache-2.0","parameters":8.2,"page_url":"https://huggingface.co/Qwen/Qwen3Guard-Gen-8B","architecture":"transformer","model_type":"chat","tagline":"**Qwen3Guard** is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV","name":"Meta-Llama-3-8B-Instruct-FP8-KV","created_by":"redhatai","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV","release_date":"2024-05-20","context_window":8192,"architecture":"transformer","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"RedHatAI/Qwen2-1.5B-Instruct-FP8","name":"Qwen2-1.5B-Instruct-FP8","created_by":"redhatai","source":"official","release_date":"2024-06-14","context_window":32768,"license":"apache-2.0","parameters":1.5,"page_url":"https://huggingface.co/RedHatAI/Qwen2-1.5B-Instruct-FP8","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"RedHatAI/Qwen2-7B-Instruct-FP8","name":"Qwen2-7B-Instruct-FP8","created_by":"redhatai","source":"official","release_date":"2024-06-14","context_window":32768,"license":"apache-2.0","parameters":7.6,"page_url":"https://huggingface.co/RedHatAI/Qwen2-7B-Instruct-FP8","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-14"},{"id":"ricdomolm/mini-coder-1.7b","name":"mini-coder-1.7b","created_by":"ricdomolm","source":"official","description":"`mini-coder-1.7b` is a 1.7B parameter model distilled from Qwen 3 Coder 30B A3B. It punches well above its weight, outperforming SWE-agent-LM 7B on [SWE-bench Verified Bash only](https://www.swebench.com/):","release_date":"2025-09-30","context_window":40960,"license":"mit","parameters":1.7,"page_url":"https://huggingface.co/ricdomolm/mini-coder-1.7b","architecture":"transformer","tagline":"`mini-coder-1.7b` is a 1.7B parameter model distilled from Qwen 3 Coder 30B A3B.","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"SakanaAI/TinySwallow-1.5B-Instruct","name":"TinySwallow-1.5B-Instruct","created_by":"sakanaai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":1.5,"page_url":"https://huggingface.co/SakanaAI/TinySwallow-1.5B-Instruct","description":"🤗 [Models](https://huggingface.co/SakanaAI) | 📚 [Paper](https://arxiv.org/abs/2501.16937) | 📝 [Blog](https://sakana.ai/taid-jp/) | 🐦 [Twitter](https://twitter.com/SakanaAILabs)","release_date":"2025-01-07","context_window":32768,"license":"apache-2.0","tagline":"🤗 [Models](https://huggingface.co/SakanaAI) | 📚 [Paper](https://arxiv.org/abs/2501.16937) | 📝 [Blog](https://sakana.ai/taid-jp/) | 🐦 [Twitter](https://twitter.com/SakanaAILabs)","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"ShahriarFerdoush/llama-3.2-1b-code-instruct","name":"llama-3.2-1b-code-instruct","created_by":"shahriarferdoush","source":"official","family":"llama-3.2","description":"A lightweight yet powerful **code-focused language model** fine-tuned from **Meta Llama-3.2-1B** using **QLoRA (4-bit)** on the **CodeAlpaca-20K** dataset. Designed for **efficient code generation, reasoning, and problem-solving** on limited GPU resources.","release_date":"2025-12-24","context_window":131072,"license":"apache-2.0","parameters":1.2,"page_url":"https://huggingface.co/ShahriarFerdoush/llama-3.2-1b-code-instruct","architecture":"transformer","model_type":"chat","tagline":"A lightweight yet powerful **code-focused language model** fine-tuned from **Meta Llama-3.2-1B** using **QLoRA (4-bit)** on the **CodeAlpaca-20K** dataset.","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-07"},{"id":"speakleash/Bielik-7B-Instruct-v0.1","name":"Bielik-7B-Instruct-v0.1","created_by":"speakleash","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"swap-uniba/LLaMAntino-3-ANITA-8B-Inst-DPO-ITA","name":"LLaMAntino-3-ANITA-8B-Inst-DPO-ITA","created_by":"swap-uniba","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/swap-uniba/LLaMAntino-3-ANITA-8B-Inst-DPO-ITA","description":"📣 New MODEL FAMILY❗ [https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA](https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA)","release_date":"2024-04-29","context_window":8192,"license":"llama3","tagline":"📣 New MODEL FAMILY❗ [https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA](https://huggingface.co/m-polignano/ANITA-NEXT-24B-Magistral-2506-VISION-ITA)","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"swiss-ai/Apertus-70B-Instruct-2509","name":"Apertus-70B-Instruct-2509","created_by":"swiss-ai","source":"official","release_date":"2025-09-01","context_window":65536,"license":"apache-2.0","parameters":71,"alias":"swiss-ai/Apertus-70B-Instruct","page_url":"https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"swiss-ai/Apertus-70B-Instruct","name":"Apertus-70B-Instruct-2509","created_by":"swiss-ai","source":"official","release_date":"2025-09-01","context_window":65536,"license":"apache-2.0","parameters":71,"page_url":"https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"snapshots":["swiss-ai/Apertus-70B-Instruct-2509"],"last_updated":"2026-06-10"},{"id":"swiss-ai/Apertus-8B-Instruct-2509","name":"Apertus-8B-Instruct-2509","created_by":"swiss-ai","source":"official","last_updated":"2026-04-21","alias":"swiss-ai/Apertus-8B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"parameters":8.1,"page_url":"https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509","release_date":"2025-08-13","context_window":65536,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"swiss-ai/Apertus-8B-Instruct","name":"Apertus-8B-Instruct-2509","created_by":"swiss-ai","source":"official","last_updated":"2026-04-21","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["swiss-ai/Apertus-8B-Instruct-2509"],"parameters":8.1,"page_url":"https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509","release_date":"2025-08-13","context_window":65536,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"]},{"id":"TinyLlama/TinyLlama-1.1B-Chat-v1.0","name":"TinyLlama-1.1B-Chat-v1.0","created_by":"tinyllama","source":"official","release_date":"2023-12-30","context_window":2048,"license":"apache-2.0","parameters":1.1,"page_url":"https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1","name":"Llama-3-Swallow-8B-Instruct-v0.1","created_by":"tokyotech-llm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1","description":"Our Swallow model has undergone continual pre-training from the [Llama 3 family](https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6), primarily with the addition of Japanese language data. The Instruct versions use supervised fine-tuning (SFT) and Chat Vector. Links to other models can be found in the index.","release_date":"2024-06-26","context_window":8192,"license":"llama3","tagline":"Our Swallow model has undergone continual pre-training from the [Llama 3 family](https://huggingface.co/collections/meta-llama/meta-llama-3-66214712577ca38149ebb2b6), primarily with the addition of Japanese language data.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.5","name":"Llama-3.1-Swallow-8B-Instruct-v0.5","created_by":"tokyotech-llm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.5","description":"Llama 3.1 Swallow is a series of large language models (8B, 70B) that were built by continual pre-training on the [Meta Llama 3.1](https://huggingface.co/collections/meta-llama/llama-31-669fc079a0c406a149a5738f) models. Llama 3.1 Swallow enhanced the Japanese language capabilities of the original Llama 3.1 while retaining the English language capabilities.","release_date":"2025-06-12","context_window":131072,"license":"llama3.3","tagline":"Llama 3.1 Swallow is a series of large language models (8B, 70B) that were built by continual pre-training on the [Meta Llama 3.1](https://huggingface.co/collections/meta-llama/llama-31-669fc079a0c406a149a5738f) models.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/gemma-2-9b-it","name":"gemma-2-9b-it","created_by":"unsloth","source":"official","last_updated":"2026-03-27","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":9.2,"page_url":"https://huggingface.co/unsloth/gemma-2-9b-it","release_date":"2024-07-03","context_window":8192,"license":"gemma","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/gemma-3-1b-it","name":"gemma-3-1b-it","created_by":"unsloth","source":"official","family":"gemma-3","description":"- Fine-tune Gemma 3 (12B) for free using our Google [Colab notebook here](https://docs.unsloth.ai/get-started/unsloth-notebooks)! - Read our Blog about Gemma 3 support: [unsloth.ai/blog/gemma3](https://unsloth.ai/blog/gemma3) - View the rest of our notebooks in our [docs here](https://docs.unsloth.ai/get-started/unsloth-notebooks). - Export your fine-tuned model to GGUF, Ollama, llama.cpp or 🤗HF.","release_date":"2025-03-12","context_window":32768,"license":"gemma","parameters":1,"page_url":"https://huggingface.co/unsloth/gemma-3-1b-it","architecture":"transformer","model_type":"chat","tagline":"- Fine-tune Gemma 3 (12B) for free using our Google [Colab notebook here](https://docs.unsloth.ai/get-started/unsloth-notebooks)! - Read our Blog about Gemma 3 support: [unsloth.ai/blog/gemma3](https://unsloth.ai/blog/gemma3) - View the rest of our notebooks in our [docs here](https://docs.unsloth.ai/get-started/unsloth-notebooks).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/llama-3-8b-Instruct","name":"llama-3-8b-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","family":"llama-3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/llama-3-8b-Instruct","description":"Directly quantized 4bit model with `bitsandbytes`.","release_date":"2024-04-18","context_window":8192,"license":"llama3","tagline":"Directly quantized 4bit model with `bitsandbytes`.","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Llama-3.1-8B-Instruct","name":"Llama-3.1-8B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/Llama-3.1-8B-Instruct","release_date":"2025-02-15","context_window":131072,"license":"llama3.1","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Llama-3.2-1B-Instruct","name":"Llama-3.2-1B-Instruct","created_by":"unsloth","source":"official","family":"Llama-3.2","release_date":"2024-09-25","context_window":131072,"license":"llama3.2","parameters":1.2,"page_url":"https://huggingface.co/unsloth/Llama-3.2-1B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Llama-3.2-3B-Instruct","name":"Llama-3.2-3B-Instruct","created_by":"unsloth","source":"official","family":"Llama-3.2","release_date":"2024-09-25","context_window":131072,"license":"llama3.2","parameters":3.2,"page_url":"https://huggingface.co/unsloth/Llama-3.2-3B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Llama-3.3-70B-Instruct","name":"Llama-3.3-70B-Instruct","created_by":"unsloth","source":"official","family":"Llama-3.3","release_date":"2024-12-06","context_window":131072,"license":"llama3.3","parameters":71,"page_url":"https://huggingface.co/unsloth/Llama-3.3-70B-Instruct","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-06"},{"id":"unsloth/Meta-Llama-3.1-8B-Instruct","name":"Meta-Llama-3.1-8B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"Llama-3.1","model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/Meta-Llama-3.1-8B-Instruct","description":"We have a free Google Colab Tesla T4 notebook for Llama 3.1 (8B) here: https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing","release_date":"2024-07-23","context_window":131072,"license":"llama3.1","tagline":"We have a free Google Colab Tesla T4 notebook for Llama 3.1 (8B) here: https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Mistral-Small-24B-Instruct-2501","name":"Mistral-Small-24B-Instruct-2501","created_by":"unsloth","source":"official","last_updated":"2026-03-24","alias":"unsloth/Mistral-Small-24B-Instruct","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":24,"page_url":"https://huggingface.co/unsloth/Mistral-Small-24B-Instruct-2501","description":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","release_date":"2025-01-30","context_window":32768,"license":"apache-2.0","tagline":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Mistral-Small-24B-Instruct","name":"Mistral-Small-24B-Instruct-2501","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"snapshots":["unsloth/Mistral-Small-24B-Instruct-2501"],"model_type":"chat","parameters":24,"page_url":"https://huggingface.co/unsloth/Mistral-Small-24B-Instruct-2501","description":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","release_date":"2025-01-30","context_window":32768,"license":"apache-2.0","tagline":"We have a free Google Colab Tesla T4 notebook for Mistral (7B) here: https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Mistral_v0.3_(7B)-Conversational.ipynb","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-0.5B-Instruct","name":"Qwen2.5-0.5B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","parameters":0.49,"page_url":"https://huggingface.co/unsloth/Qwen2.5-0.5B-Instruct","architecture":"transformer","model_type":"chat","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen2.5-1.5B-Instruct","name":"Qwen2.5-1.5B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","parameters":1.5,"page_url":"https://huggingface.co/unsloth/Qwen2.5-1.5B-Instruct","architecture":"transformer","model_type":"chat","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen2.5-14B-Instruct","name":"Qwen2.5-14B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":15,"page_url":"https://huggingface.co/unsloth/Qwen2.5-14B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-23","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-32B-Instruct","name":"Qwen2.5-32B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":33,"page_url":"https://huggingface.co/unsloth/Qwen2.5-32B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-23","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-3B-Instruct","name":"Qwen2.5-3B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"other","parameters":3.1,"page_url":"https://huggingface.co/unsloth/Qwen2.5-3B-Instruct","architecture":"transformer","model_type":"chat","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen2.5-7B-Instruct","name":"Qwen2.5-7B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7.6,"page_url":"https://huggingface.co/unsloth/Qwen2.5-7B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-18","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-Coder-32B-Instruct","name":"Qwen2.5-Coder-32B-Instruct","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","parameters":33,"page_url":"https://huggingface.co/unsloth/Qwen2.5-Coder-32B-Instruct","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-11-12","context_window":32768,"license":"apache-2.0","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/Qwen2.5-Coder-7B-Instruct","name":"Qwen2.5-Coder-7B-Instruct","created_by":"unsloth","source":"official","description":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing). Also a [Qwen 2.5 conversational style notebook](https://colab.research.google.com/drive/1qN1CEalC70EO1wGKhNxs1go1W9So61R5?usp=sharing).","release_date":"2024-09-23","context_window":32768,"license":"apache-2.0","parameters":7.6,"page_url":"https://huggingface.co/unsloth/Qwen2.5-Coder-7B-Instruct","architecture":"transformer","model_type":"code","tagline":"We have a Qwen 2.5 (all model sizes) [free Google Colab Tesla T4 notebook](https://colab.research.google.com/drive/1Kose-ucXO1IBaZq5BvbwWieuubP7hxvQ?usp=sharing).","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-16"},{"id":"unsloth/Qwen3-1.7B-Base","name":"Qwen3-1.7B-Base","created_by":"unsloth","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-1.7B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-17"},{"id":"unsloth/Qwen3-4B-Base","name":"Qwen3-4B-Base","created_by":"unsloth","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-4B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02"},{"id":"unsloth/Qwen3-4B-Instruct-2507","name":"Qwen3-4B-Instruct-2507","created_by":"unsloth","source":"official","release_date":"2025-08-06","context_window":262144,"model_type":"chat","license":"apache-2.0","alias":"unsloth/Qwen3-4B-Instruct","page_url":"https://huggingface.co/unsloth/Qwen3-4B-Instruct-2507","architecture":"transformer","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen3-4B-Instruct","name":"Qwen3-4B-Instruct-2507","created_by":"unsloth","source":"official","release_date":"2025-08-06","context_window":262144,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-4B-Instruct-2507","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"snapshots":["unsloth/Qwen3-4B-Instruct-2507"],"last_updated":"2026-05-01"},{"id":"unsloth/Qwen3-8B-Base","name":"Qwen3-8B-Base","created_by":"unsloth","source":"official","release_date":"2025-04-28","context_window":32768,"license":"apache-2.0","page_url":"https://huggingface.co/unsloth/Qwen3-8B-Base","architecture":"transformer","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-03-26"},{"id":"unsloth/Qwen3-8B","name":"Qwen3-8B","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"page_url":"https://huggingface.co/unsloth/Qwen3-8B","release_date":"2025-04-28","context_window":40960,"license":"apache-2.0","architecture":"transformer","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"unsloth/tinyllama-chat","name":"tinyllama-chat","created_by":"unsloth","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"page_url":"https://huggingface.co/unsloth/tinyllama-chat","description":"[](https://discord.gg/u54VK8m8tk) [](https://ko-fi.com/unsloth) [](https://github.com/unslothai/unsloth)","release_date":"2024-02-14","context_window":2048,"license":"apache-2.0","parameters":1.1,"tagline":"[](https://discord.gg/u54VK8m8tk) [Discord community.","release_date":"2025-07-20","context_window":131072,"license":"mit","tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.5-Air","name":"GLM-4.5-Air","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.5","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.5-Air","description":"👋 Join our Discord community.","release_date":"2025-07-20","context_window":131072,"license":"mit","parameters":110,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.5","name":"GLM-4.5","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.5","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.5","description":"👋 Join our Discord community.","release_date":"2025-07-20","context_window":131072,"license":"mit","parameters":358,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.6-FP8","name":"GLM-4.6-FP8","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.6","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.6-FP8","description":"👋 Join our Discord community.","release_date":"2025-09-29","context_window":202752,"license":"mit","tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.6","name":"GLM-4.6","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.6","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.6","description":"👋 Join our Discord community.","release_date":"2025-09-29","context_window":202752,"license":"mit","parameters":357,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.7-Flash","name":"GLM-4.7-Flash","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.7","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.7-Flash","description":"👋 Join our Discord community.","release_date":"2026-01-19","context_window":202752,"license":"mit","parameters":31,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-4.7-FP8","name":"GLM-4.7-FP8","created_by":"zhipu","source":"official","last_updated":"2026-05-29","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.7","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.7-FP8","description":"👋 Join our Discord community.","release_date":"2025-12-22","context_window":202752,"license":"mit","tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"parameters":358},{"id":"zai-org/GLM-4.7","name":"GLM-4.7","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-4.7","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-4.7","description":"👋 Join our Discord community.","release_date":"2025-12-22","context_window":202752,"license":"mit","parameters":358,"tagline":"👋 Join our Discord community.","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-5.1-FP8","name":"GLM-5.1-FP8","created_by":"zhipu","source":"official","family":"GLM-5.1","release_date":"2026-04-03","context_window":202752,"license":"mit","parameters":754,"page_url":"https://huggingface.co/zai-org/GLM-5.1-FP8","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-10"},{"id":"zai-org/GLM-5.1","name":"GLM-5.1","created_by":"zhipu","source":"official","family":"GLM-5.1","release_date":"2026-04-03","context_window":202752,"license":"mit","parameters":754,"page_url":"https://huggingface.co/zai-org/GLM-5.1","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-12"},{"id":"zai-org/GLM-5.2-FP8","name":"GLM-5.2-FP8","created_by":"zhipu","source":"official","family":"GLM-5.2","release_date":"2026-06-16","context_window":1048576,"license":"mit","parameters":753,"page_url":"https://huggingface.co/zai-org/GLM-5.2-FP8","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"zai-org/GLM-5.2","name":"GLM-5.2","created_by":"zhipu","source":"official","family":"GLM-5.2","release_date":"2026-06-16","context_window":1048576,"license":"mit","parameters":753,"page_url":"https://huggingface.co/zai-org/GLM-5.2","architecture":"moe","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"zai-org/GLM-5","name":"GLM-5","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"GLM-5","model_type":"chat","page_url":"https://huggingface.co/zai-org/GLM-5","description":"👋 Join our WeChat or Discord community.","release_date":"2026-02-11","context_window":202752,"license":"mit","parameters":754,"tagline":"👋 Join our WeChat or Discord","architecture":"moe","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai-org/GLM-Z1-32B-0414","name":"GLM-Z1-32B-0414","created_by":"zhipu","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"chat","family":"glm","parameters":32}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/mistral.ts b/packages/npm/src/providers/mistral.ts
index 4311c9c0d..a6e241832 100644
--- a/packages/npm/src/providers/mistral.ts
+++ b/packages/npm/src/providers/mistral.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"mistral","name":"Mistral AI","region":"FR","headquarters":"Paris, Île-de-France","founded":2023,"url":"https://mistral.ai","api_url":"https://api.mistral.ai/v1","docs_url":"https://docs.mistral.ai","pricing_url":"https://mistral.ai/products","description":"European AI company building open and commercial language models optimized for efficiency.","type":"direct","playground_url":"https://console.mistral.ai","status_url":"https://status.mistral.ai","changelog_url":"https://docs.mistral.ai/getting-started/changelog/","sdk":{"python":"mistralai","javascript":"@mistralai/mistralai"},"openai_compatible":true,"free_tier":true,"github_url":"https://github.com/mistralai","models_url":"https://docs.mistral.ai/getting-started/models/","twitter_url":"https://x.com/MistralAI","discord_url":"https://discord.com/invite/mistralai","blog_url":"https://mistral.ai/news","terms_url":"https://mistral.ai/terms-of-use","support_url":"https://help.mistral.ai","icon":"","models":[{"id":"codestral-2405","name":"codestral-2405","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"deprecated","deprecation_date":"2024-12-02","context_window":32000,"alias":"codestral","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"code","pricing":{"input":0.3,"output":0.9},"license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-24-05","open_weight":true,"description":"Our first cutting-edge language model for coding and FIM, released under the MNPL license.","tagline":"Our first cutting-edge language model for coding and FIM, released under the MNPL license."},{"id":"codestral-2501","name":"codestral-2501","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"deprecated","deprecation_date":"2025-11-06","context_window":128000,"alias":"codestral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"code","pricing":{"input":0.3,"output":0.9},"license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-25-01","open_weight":true,"description":"Our cutting-edge language model for coding. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.","tagline":"Our cutting-edge language model for coding.","tools":["function_calling"]},{"id":"codestral-2508","name":"codestral-2508","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":128000,"alias":"codestral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":0.9},"model_type":"code","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-25-08","open_weight":true,"description":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","tagline":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","tools":["function_calling"]},{"id":"codestral-embed-2505","name":"codestral-embed-2505","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":8000,"alias":"codestral-embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"embed","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-embed-25-05","open_weight":true,"tools":["function_calling"]},{"id":"codestral-embed","name":"codestral-embed","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":8000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["codestral-embed-2505"],"model_type":"embed","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-embed-25-05","open_weight":true,"tools":["function_calling"]},{"id":"codestral-mamba-7b-01","name":"codestral-mamba-7b-01","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":256000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"code","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/codestral-mamba-7b-0-1","open_weight":true,"description":"A coding dedicated model using a mamba 2 architecture for research purposes under an open license.","tagline":"A coding dedicated model using a mamba 2 architecture for research purposes under an open license."},{"id":"codestral","name":"codestral","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.9},"snapshots":["codestral-2508","codestral-2501","codestral-2405"],"model_type":"code","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-24-05","open_weight":true,"description":"Our first cutting-edge language model for coding and FIM, released under the MNPL license.","tagline":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","tools":["function_calling"]},{"id":"devstral-medium-2507","name":"devstral-medium-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"alias":"devstral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"code","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/devstral-medium-1-0-25-07","open_weight":false,"description":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"devstral-medium-latest","name":"devstral-medium-latest","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"active","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"code","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/devstral-2-25-12","open_weight":false,"description":"Our frontier code agents model for solving software engineering tasks; excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our frontier code agents model for solving software engineering tasks; excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"devstral-medium","name":"devstral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"snapshots":["devstral-medium-2507"],"model_type":"code","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/devstral-medium-1-0-25-07","open_weight":false,"description":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"devstral-small-2505","name":"devstral-small-2505","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","deprecation_date":"2025-10-31","context_window":128000,"alias":"devstral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"code","pricing":{"input":0.1,"output":0.3},"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-1-0-25-05","open_weight":true,"description":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"devstral-small-2507","name":"devstral-small-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"alias":"devstral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"model_type":"code","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-1-1-25-07","open_weight":true,"description":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"devstral-small-2512","name":"devstral-small-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"active","context_window":256000,"alias":"devstral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"model_type":"code","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-2-25-12","open_weight":true,"description":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"devstral-small","name":"devstral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"snapshots":["devstral-small-2512","devstral-small-2507","devstral-small-2505"],"model_type":"code","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-1-0-25-05","open_weight":true,"description":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"leanstral-2603","name":"leanstral-2603","created_by":"mistral","source":"official","description":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories. 119B parameters with 6.5B active.","status":"active","context_window":256000,"license":"apache-2.0","alias":"leanstral","page_url":"https://docs.mistral.ai/models/model-cards/leanstral-26-03","tagline":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"leanstral","name":"leanstral","created_by":"mistral","source":"official","description":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories. 119B parameters with 6.5B active.","status":"active","context_window":256000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/leanstral-26-03","tagline":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["leanstral-2603"],"last_updated":"2026-06-03"},{"id":"magistral-medium-2506","name":"magistral-medium-2506","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-0-25-06","open_weight":false,"description":"Our first frontier-class reasoning model released June 2025.","tagline":"Our first frontier-class reasoning model released June 2025.","tools":["function_calling"]},{"id":"magistral-medium-2507","name":"magistral-medium-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-1-25-07","open_weight":false,"description":"Our frontier-class reasoning model released July 2025.","tagline":"Our frontier-class reasoning model released July 2025.","tools":["function_calling"]},{"id":"magistral-medium-2509","name":"magistral-medium-2509","created_by":"mistral","source":"official","last_updated":"2026-06-18","status":"deprecated","context_window":128000,"alias":"magistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-2-25-09","open_weight":false,"description":"Our frontier-class multimodal reasoning model update of September 2025.","tagline":"Our frontier-class multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"magistral-medium","name":"magistral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"deprecated","context_window":40000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"snapshots":["magistral-medium-2509","magistral-medium-2507","magistral-medium-2506"],"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-0-25-06","open_weight":false,"description":"Our first frontier-class reasoning model released June 2025.","tagline":"Our frontier-class multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"magistral-small-2506","name":"magistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-0-25-06","open_weight":true,"description":"Our small reasoning model released July 2025.","tagline":"Our small reasoning model released July 2025.","tools":["function_calling"]},{"id":"magistral-small-2507","name":"magistral-small-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-1-25-07","open_weight":true,"description":"Our small reasoning model released July 2025.","tagline":"Our small reasoning model released July 2025.","tools":["function_calling"]},{"id":"magistral-small-2509","name":"magistral-small-2509","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":128000,"alias":"magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09","open_weight":true,"description":"Our small multimodal reasoning model update of September 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"magistral-small","name":"magistral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":40000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"snapshots":["magistral-small-2509","magistral-small-2507","magistral-small-2506"],"deprecation_date":"2025-10-31","model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-0-25-06","open_weight":true,"description":"Our small reasoning model released July 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"mathstral-7b-01","name":"mathstral-7b-01","created_by":"mistral","source":"official","description":"A math dedicated model for research purposes under an open license.","status":"active","context_window":32000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mathstral-7b-0-1","tagline":"A math dedicated model for research purposes under an open license.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"ministral-14b-2512","name":"ministral-14b-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"alias":"ministral-14b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"model_type":"chat","license":"apache-2.0","parameters":14,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-14b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-14b","name":"ministral-14b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"snapshots":["ministral-14b-2512"],"model_type":"chat","license":"apache-2.0","parameters":14,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-14b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-3b-2410","name":"ministral-3b-2410","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"alias":"ministral-3b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"ministral-3b-2512","name":"ministral-3b-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"alias":"ministral-3b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-3b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-3b","name":"ministral-3b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"snapshots":["ministral-3b-2512","ministral-3b-2410"],"model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"ministral-8b-2410","name":"ministral-8b-2410","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"alias":"ministral-8b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","license":"apache-2.0","parameters":8,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-8b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"ministral-8b-2512","name":"ministral-8b-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"alias":"ministral-8b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","license":"apache-2.0","parameters":8,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-8b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-8b","name":"ministral-8b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"snapshots":["ministral-8b-2512","ministral-8b-2410"],"model_type":"chat","license":"apache-2.0","parameters":8,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-8b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"mistral-7b-01","name":"mistral-7b-01","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":8000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-7b","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/mistral-7b-0-1","open_weight":true,"description":"Our very first open model, SOTA competing against models up to 2 to 3x larger.","tagline":"Our very first open model, SOTA competing against models up to 2 to 3x larger."},{"id":"mistral-7b-02","name":"mistral-7b-02","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-7b","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/mistral-7b-0-2","open_weight":true,"description":"An update to our first open model with better context.","tagline":"An update to our first open model with better context."},{"id":"mistral-7b-03","name":"mistral-7b-03","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-7b","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/mistral-7b-0-3","open_weight":true,"description":"A minor update to our first open model.","tagline":"A minor update to our first open model."},{"id":"mistral-embed-2312","name":"mistral-embed-2312","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":8000,"alias":"mistral-embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"embed","family":"mistral-embed","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-embed-23-12","open_weight":true,"tools":["function_calling"]},{"id":"mistral-embed","name":"mistral-embed","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":8000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["mistral-embed-2312"],"model_type":"embed","family":"mistral-embed","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-embed-23-12","open_weight":true,"tools":["function_calling"]},{"id":"mistral-large-2402","name":"mistral-large-2402","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","deprecation_date":"2024-11-30","context_window":32000,"alias":"mistral-large","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-1-0-24-02","open_weight":true,"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities."},{"id":"mistral-large-2407","name":"mistral-large-2407","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","deprecation_date":"2024-11-30","context_window":128000,"alias":"mistral-large","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-2-0-24-07","open_weight":true,"description":"A powerfull model with open weights under a MRL license, competing agains SOTA much larger models.","tagline":"A powerfull model with open weights under a MRL license, competing agains SOTA much larger models."},{"id":"mistral-large-2411","name":"mistral-large-2411","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","context_window":128000,"alias":"mistral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-2-1-24-11","open_weight":true,"description":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024.","tagline":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"mistral-large-2512","name":"mistral-large-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"active","context_window":256000,"alias":"mistral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-3-25-12","open_weight":true,"tools":["function_calling"],"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities."},{"id":"mistral-large","name":"mistral-large","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"snapshots":["mistral-large-2512","mistral-large-2411","mistral-large-2407","mistral-large-2402"],"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-1-0-24-02","open_weight":true,"tools":["function_calling"],"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024."},{"id":"mistral-medium-2312","name":"mistral-medium-2312","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-medium","status":"deprecated","deprecation_date":"2024-11-30","context_window":32000,"alias":"mistral-medium","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-1-0-23-12","open_weight":false,"description":"Our first SOTA enterprise grade model.","tagline":"Our first SOTA enterprise grade model."},{"id":"mistral-medium-2505","name":"mistral-medium-2505","created_by":"mistral","source":"official","last_updated":"2026-06-18","family":"mistral-medium","status":"deprecated","context_window":128000,"alias":"mistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-25-05","open_weight":false,"description":"Our frontier-class multimodal model released August 2025. Improving tone and performance.","tagline":"Our frontier-class multimodal model released August 2025.","tools":["function_calling"]},{"id":"mistral-medium-2508","name":"mistral-medium-2508","created_by":"mistral","source":"official","last_updated":"2026-06-18","family":"mistral-medium","status":"deprecated","context_window":128000,"alias":"mistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-1-25-08","open_weight":false,"description":"Our frontier-class multimodal model released August 2025. Improving tone and performance.","tagline":"Our frontier-class multimodal model released August 2025.","tools":["function_calling"]},{"id":"mistral-medium-3-5-2604","name":"mistral-medium-3-5-2604","created_by":"mistral","source":"official","family":"mistral-medium","description":"Our frontier-class multimodal model optimized for agentic and coding use cases. Released as open weights under a Modified MIT license.","status":"active","context_window":256000,"model_type":"chat","license":"proprietary","alias":"mistral-medium-3-5","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","tagline":"Our frontier-class multimodal model optimized for agentic and coding use cases.","capabilities":{"streaming":true,"tool_call":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"open_weight":false,"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"mistral-medium-3-5","name":"mistral-medium-3-5","created_by":"mistral","source":"official","family":"mistral-medium","description":"Our frontier-class multimodal model optimized for agentic and coding use cases. Released as open weights under a Modified MIT license.","status":"active","context_window":256000,"license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","model_type":"chat","tagline":"Our frontier-class multimodal model optimized for agentic and coding use cases.","capabilities":{"streaming":true,"tool_call":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"open_weight":false,"tools":["function_calling"],"snapshots":["mistral-medium-3-5-2604"],"last_updated":"2026-06-03"},{"id":"mistral-medium-latest","name":"mistral-medium-latest","created_by":"mistral","source":"official","family":"mistral-medium","description":"Our frontier-class multimodal model optimized for agentic and coding use cases. Released as open weights under a Modified MIT license.","status":"active","context_window":256000,"license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","model_type":"chat","tagline":"Our frontier-class multimodal model optimized for agentic and coding use cases.","capabilities":{"streaming":true,"tool_call":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"open_weight":false,"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"mistral-medium","name":"mistral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-10","family":"mistral-medium","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2},"snapshots":["mistral-medium-2508","mistral-medium-2505","mistral-medium-2312"],"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-1-0-23-12","open_weight":false,"description":"Our first SOTA enterprise grade model.","tagline":"Our frontier-class multimodal model released August 2025.","tools":["function_calling"]},{"id":"mistral-moderation-2411","name":"mistral-moderation-2411","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":8000,"alias":"mistral-moderation","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"moderation","family":"mistral-moderation","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-moderation-24-11","open_weight":true,"description":"Our latest moderation model with 128k context window and jailbreaking detection. Strong performance on complex multilingual data including long multi-turn conversations.","tagline":"Our latest moderation model with 128k context window and jailbreaking detection, based on Ministral 3B.","tools":["function_calling"],"deprecation_date":"2026-03-31"},{"id":"mistral-moderation-2603","name":"mistral-moderation-2603","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":128000,"alias":"mistral-moderation","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"moderation","family":"mistral-moderation","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-moderation-26-03","open_weight":true,"description":"Our latest moderation model with 128k context window and jailbreaking detection. Strong performance on complex multilingual data including long multi-turn conversations.","tagline":"Our latest moderation model with 128k context window and jailbreaking detection, based on Ministral 3B.","tools":["function_calling"]},{"id":"mistral-moderation","name":"mistral-moderation","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":8000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["mistral-moderation-2603","mistral-moderation-2411"],"model_type":"moderation","family":"mistral-moderation","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-moderation-24-11","open_weight":true,"description":"Our latest moderation model with 128k context window and jailbreaking detection. Strong performance on complex multilingual data including long multi-turn conversations.","tagline":"Our latest moderation model with 128k context window and jailbreaking detection, based on Ministral 3B.","tools":["function_calling"]},{"id":"mistral-nemo-2407","name":"mistral-nemo-2407","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":128000,"alias":"mistral-nemo","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","family":"mistral-nemo","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-nemo-12b-24-07","open_weight":true,"description":"Our best multilingual open source model released July 2024.","tagline":"Our best multilingual open source model released July 2024.","tools":["function_calling"]},{"id":"mistral-nemo","name":"mistral-nemo","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"snapshots":["mistral-nemo-2407"],"model_type":"chat","family":"mistral-nemo","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-nemo-12b-24-07","open_weight":true,"description":"Our best multilingual open source model released July 2024.","tagline":"Our best multilingual open source model released July 2024.","tools":["function_calling"]},{"id":"mistral-next","name":"mistral-next","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-next","open_weight":true,"description":"A prototype model, designed to be brief and concise.","tagline":"A prototype model, designed to be brief and concise."},{"id":"mistral-ocr-2503","name":"mistral-ocr-2503","created_by":"mistral","source":"official","description":"Our first OCR optimized model.","status":"deprecated","model_type":"chat","license":"apache-2.0","alias":"mistral-ocr","page_url":"https://docs.mistral.ai/models/model-cards/ocr-25-03","tagline":"Our first OCR optimized model.","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"mistral-ocr-2505","name":"mistral-ocr-2505","created_by":"mistral","source":"official","status":"deprecated","model_type":"chat","license":"apache-2.0","alias":"mistral-ocr","page_url":"https://docs.mistral.ai/models/model-cards/ocr-2-25-05","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-03","description":"Our first OCR optimized model.","tagline":"Our first OCR optimized model."},{"id":"mistral-ocr-2512","name":"mistral-ocr-2512","created_by":"mistral","source":"official","status":"active","model_type":"chat","license":"apache-2.0","alias":"mistral-ocr","page_url":"https://docs.mistral.ai/models/model-cards/ocr-3-25-12","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-03","description":"Our first OCR optimized model.","tagline":"Our first OCR optimized model."},{"id":"mistral-ocr","name":"mistral-ocr","created_by":"mistral","source":"official","status":"deprecated","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/ocr-25-03","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["mistral-ocr-2512","mistral-ocr-2505","mistral-ocr-2503"],"last_updated":"2026-06-03","description":"Our first OCR optimized model.","tagline":"Our first OCR optimized model."},{"id":"mistral-saba-2502","name":"mistral-saba-2502","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-06-10","context_window":32000,"alias":"mistral-saba","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-saba","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-saba-25-02","open_weight":true,"description":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia.","tagline":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia."},{"id":"mistral-saba","name":"mistral-saba","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-06-10","context_window":32000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"snapshots":["mistral-saba-2502"],"model_type":"chat","family":"mistral-saba","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-saba-25-02","open_weight":true,"description":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia.","tagline":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia."},{"id":"mistral-small-2402","name":"mistral-small-2402","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2024-11-30","context_window":32000,"alias":"mistral-small","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-1-0-24-02","open_weight":true,"description":"An enterprise grade small model for efficiency.","tagline":"An enterprise grade small model for efficiency."},{"id":"mistral-small-2409","name":"mistral-small-2409","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2025-11-06","context_window":32000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-2-0-24-09","open_weight":true,"description":"An update to our efficient small model with open weights under an MRL license.","tagline":"An update to our efficient small model with open weights under an MRL license.","tools":["function_calling"]},{"id":"mistral-small-2501","name":"mistral-small-2501","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2025-11-06","context_window":32000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-3-0-25-01","open_weight":true,"description":"A small efficient and powerfull 24B open model for personal or commercial use.","tagline":"A small efficient and powerfull 24B open model for personal or commercial use.","tools":["function_calling"]},{"id":"mistral-small-2503","name":"mistral-small-2503","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2025-11-06","context_window":128000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-3-1-25-03","open_weight":true,"description":"A new leader in the small models category with image understanding capabilities, released March 2025.","tagline":"A new leader in the small models category with image understanding capabilities, released March 2025.","tools":["function_calling"]},{"id":"mistral-small-2506","name":"mistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","context_window":128000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-3-2-25-06","open_weight":true,"description":"An update to our previous small model, released June 2025.","tagline":"An update to our previous small model, released June 2025.","tools":["function_calling"]},{"id":"mistral-small-2603","name":"mistral-small-2603","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"active","context_window":256000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-4-0-26-03","open_weight":true,"description":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model. 119B parameters with 6.5B active.","tagline":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model.","tools":["function_calling"]},{"id":"mistral-small-creative-2512","name":"mistral-small-creative-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"active","context_window":32000,"alias":"mistral-small-creative","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-creative-25-12","open_weight":true,"description":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction following and conversational agents.","tagline":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction fol","tools":["function_calling"]},{"id":"mistral-small-creative","name":"mistral-small-creative","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"active","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"snapshots":["mistral-small-creative-2512"],"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-creative-25-12","open_weight":true,"description":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction following and conversational agents.","tagline":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction fol","tools":["function_calling"]},{"id":"mistral-small","name":"mistral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"snapshots":["mistral-small-2603","mistral-small-2506","mistral-small-2503","mistral-small-2501","mistral-small-2409","mistral-small-2402"],"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-1-0-24-02","open_weight":true,"description":"An enterprise grade small model for efficiency.","tagline":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model.","tools":["function_calling"]},{"id":"mixtral-8x22b-0-1-03","name":"mixtral-8x22b-0-1-03","created_by":"mistral","source":"official","family":"mixtral","description":"A large open MoE model competing with much larger dense models.","status":"active","context_window":64000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mixtral-8x22b-0-1-0-3","model_type":"chat","tagline":"A large open MoE model competing with much larger dense models.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"mixtral-8x7b-01","name":"mixtral-8x7b-01","created_by":"mistral","source":"official","family":"mixtral","description":"Our first open MoE model, SOTA competing against models up to 2 to 3x larger.","status":"active","context_window":32000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mixtral-8x7b-0-1","model_type":"chat","tagline":"Our first open MoE model, SOTA competing against models up to 2 to 3x larger.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"pixtral-12b-2409","name":"pixtral-12b-2409","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"alias":"pixtral-12b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"chat","license":"apache-2.0","parameters":12,"page_url":"https://docs.mistral.ai/models/model-cards/pixtral-12b-24-09","open_weight":true,"description":"A 12B model with image understanding capabilities in addition to text.","tagline":"A 12B model with image understanding capabilities in addition to text.","tools":["function_calling"]},{"id":"pixtral-12b","name":"pixtral-12b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["pixtral-12b-2409"],"model_type":"chat","license":"apache-2.0","parameters":12,"page_url":"https://docs.mistral.ai/models/model-cards/pixtral-12b-24-09","open_weight":true,"description":"A 12B model with image understanding capabilities in addition to text.","tagline":"A 12B model with image understanding capabilities in addition to text.","tools":["function_calling"]},{"id":"pixtral-large-2411","name":"pixtral-large-2411","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","context_window":128000,"alias":"pixtral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":6},"model_type":"chat","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/pixtral-large-24-11","open_weight":true,"tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"pixtral-large","name":"pixtral-large","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":6},"snapshots":["pixtral-large-2411"],"model_type":"chat","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/pixtral-large-24-11","open_weight":true,"tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"voxtral-mini-2507","name":"voxtral-mini-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":32000,"alias":"voxtral-mini","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-25-07","open_weight":true,"description":"A mini version of our first audio input model.","tagline":"An efficient audio input model, fine-tuned and optimized for transcription purposes only.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"voxtral-mini-2602","name":"voxtral-mini-2602","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"active","alias":"voxtral-mini","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"context_window":32000,"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-transcribe-26-02","open_weight":true,"description":"An efficient audio input model, pre-trained and optimized for transcription purposes only.","tagline":"An efficient audio input model, fine-tuned and optimized for transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-mini-transcribe-realtime-2602","name":"voxtral-mini-transcribe-realtime-2602","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"active","alias":"voxtral-mini-transcribe-realtime","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"transcription","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-transcribe-realtime-26-02","open_weight":true,"description":"An efficient audio input model, pre-trained and optimized for live transcription purposes only.","tagline":"An efficient audio input model, fine-tuned and optimized for live transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-mini-transcribe-realtime","name":"voxtral-mini-transcribe-realtime","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["voxtral-mini-transcribe-realtime-2602"],"model_type":"transcription","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-transcribe-realtime-26-02","open_weight":true,"description":"An efficient audio input model, pre-trained and optimized for live transcription purposes only.","tagline":"An efficient audio input model, fine-tuned and optimized for live transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-mini-tts-2603","name":"voxtral-mini-tts-2603","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~90ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","model_type":"tts","license":"apache-2.0","alias":"voxtral-mini-tts","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"voxtral-mini-tts","name":"voxtral-mini-tts","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~90ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03","model_type":"tts","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["voxtral-mini-tts-2603"],"last_updated":"2026-06-03"},{"id":"voxtral-mini","name":"voxtral-mini","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["voxtral-mini-2507","voxtral-mini-2602"],"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-25-07","open_weight":true,"description":"A mini version of our first audio input model.","tagline":"An efficient audio input model, fine-tuned and optimized for transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-small-2507","name":"voxtral-small-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"alias":"voxtral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-small-25-07","open_weight":true,"description":"Our first model with audio input capabilities for instruct use cases.","tagline":"Our first model with audio input capabilities for instruct use cases.","tools":["function_calling"]},{"id":"voxtral-small","name":"voxtral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["voxtral-small-2507"],"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-small-25-07","open_weight":true,"description":"Our first model with audio input capabilities for instruct use cases.","tagline":"Our first model with audio input capabilities for instruct use cases.","tools":["function_calling"]},{"id":"voxtral-tts-2603","name":"voxtral-tts-2603","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~100ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","model_type":"tts","license":"apache-2.0","alias":"voxtral-tts","page_url":"https://docs.mistral.ai/models/voxtral-tts-26-03","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-03-27"},{"id":"voxtral-tts","name":"voxtral-tts","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~100ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/voxtral-tts-26-03","model_type":"tts","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["voxtral-tts-2603"],"last_updated":"2026-03-27"}]};
+export const provider: ProviderWithModels = {"id":"mistral","name":"Mistral AI","region":"FR","headquarters":"Paris, Île-de-France","founded":2023,"url":"https://mistral.ai","api_url":"https://api.mistral.ai/v1","docs_url":"https://docs.mistral.ai","pricing_url":"https://mistral.ai/products","description":"European AI company building open and commercial language models optimized for efficiency.","type":"direct","playground_url":"https://console.mistral.ai","status_url":"https://status.mistral.ai","changelog_url":"https://docs.mistral.ai/getting-started/changelog/","sdk":{"python":"mistralai","javascript":"@mistralai/mistralai"},"openai_compatible":true,"free_tier":true,"github_url":"https://github.com/mistralai","models_url":"https://docs.mistral.ai/getting-started/models/","twitter_url":"https://x.com/MistralAI","discord_url":"https://discord.com/invite/mistralai","blog_url":"https://mistral.ai/news","terms_url":"https://mistral.ai/terms-of-use","support_url":"https://help.mistral.ai","icon":"","models":[{"id":"codestral-2405","name":"codestral-2405","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"deprecated","deprecation_date":"2024-12-02","context_window":32000,"alias":"codestral","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"code","pricing":{"input":0.3,"output":0.9},"license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-24-05","open_weight":true,"description":"Our first cutting-edge language model for coding and FIM, released under the MNPL license.","tagline":"Our first cutting-edge language model for coding and FIM, released under the MNPL license."},{"id":"codestral-2501","name":"codestral-2501","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"deprecated","deprecation_date":"2025-11-06","context_window":128000,"alias":"codestral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"code","pricing":{"input":0.3,"output":0.9},"license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-25-01","open_weight":true,"description":"Our cutting-edge language model for coding. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.","tagline":"Our cutting-edge language model for coding.","tools":["function_calling"]},{"id":"codestral-2508","name":"codestral-2508","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":128000,"alias":"codestral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":0.9},"model_type":"code","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-25-08","open_weight":true,"description":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","tagline":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","tools":["function_calling"]},{"id":"codestral-embed-2505","name":"codestral-embed-2505","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":8000,"alias":"codestral-embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"embed","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-embed-25-05","open_weight":true,"tools":["function_calling"]},{"id":"codestral-embed","name":"codestral-embed","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":8000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["codestral-embed-2505"],"model_type":"embed","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-embed-25-05","open_weight":true,"tools":["function_calling"]},{"id":"codestral-mamba-7b-01","name":"codestral-mamba-7b-01","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"active","context_window":256000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"code","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/codestral-mamba-7b-0-1","open_weight":true,"description":"A coding dedicated model using a mamba 2 architecture for research purposes under an open license.","tagline":"A coding dedicated model using a mamba 2 architecture for research purposes under an open license."},{"id":"codestral","name":"codestral","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"codestral","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.9},"snapshots":["codestral-2508","codestral-2501","codestral-2405"],"model_type":"code","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/codestral-24-05","open_weight":true,"description":"Our first cutting-edge language model for coding and FIM, released under the MNPL license.","tagline":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","tools":["function_calling"]},{"id":"devstral-medium-2507","name":"devstral-medium-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"alias":"devstral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"code","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/devstral-medium-1-0-25-07","open_weight":false,"description":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"devstral-medium-latest","name":"devstral-medium-latest","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"active","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"code","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/devstral-2-25-12","open_weight":false,"description":"Our frontier code agents model for solving software engineering tasks; excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our frontier code agents model for solving software engineering tasks; excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"devstral-medium","name":"devstral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"snapshots":["devstral-medium-2507"],"model_type":"code","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/devstral-medium-1-0-25-07","open_weight":false,"description":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"An enterprise grade text model, that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"devstral-small-2505","name":"devstral-small-2505","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","deprecation_date":"2025-10-31","context_window":128000,"alias":"devstral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"code","pricing":{"input":0.1,"output":0.3},"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-1-0-25-05","open_weight":true,"description":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"devstral-small-2507","name":"devstral-small-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"alias":"devstral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"model_type":"code","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-1-1-25-07","open_weight":true,"description":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"devstral-small-2512","name":"devstral-small-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"active","context_window":256000,"alias":"devstral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"model_type":"code","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-2-25-12","open_weight":true,"description":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"devstral-small","name":"devstral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"devstral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"snapshots":["devstral-small-2512","devstral-small-2507","devstral-small-2505"],"model_type":"code","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/devstral-small-1-0-25-05","open_weight":true,"description":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tools":["function_calling"]},{"id":"leanstral-2603","name":"leanstral-2603","created_by":"mistral","source":"official","description":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories. 119B parameters with 6.5B active.","status":"active","context_window":256000,"license":"apache-2.0","alias":"leanstral","page_url":"https://docs.mistral.ai/models/model-cards/leanstral-26-03","tagline":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"leanstral","name":"leanstral","created_by":"mistral","source":"official","description":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories. 119B parameters with 6.5B active.","status":"active","context_window":256000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/leanstral-26-03","tagline":"Our first open-source code agent designed for Lean 4, built for formal proof engineering in realistic repositories.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["leanstral-2603"],"last_updated":"2026-06-03"},{"id":"magistral-medium-2506","name":"magistral-medium-2506","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-0-25-06","open_weight":false,"description":"Our first frontier-class reasoning model released June 2025.","tagline":"Our first frontier-class reasoning model released June 2025.","tools":["function_calling"]},{"id":"magistral-medium-2507","name":"magistral-medium-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-1-25-07","open_weight":false,"description":"Our frontier-class reasoning model released July 2025.","tagline":"Our frontier-class reasoning model released July 2025.","tools":["function_calling"]},{"id":"magistral-medium-2509","name":"magistral-medium-2509","created_by":"mistral","source":"official","last_updated":"2026-06-18","status":"deprecated","context_window":128000,"alias":"magistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-2-25-09","open_weight":false,"description":"Our frontier-class multimodal reasoning model update of September 2025.","tagline":"Our frontier-class multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"magistral-medium","name":"magistral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"deprecated","context_window":40000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":5},"snapshots":["magistral-medium-2509","magistral-medium-2507","magistral-medium-2506"],"model_type":"reasoning","family":"magistral","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/magistral-medium-1-0-25-06","open_weight":false,"description":"Our first frontier-class reasoning model released June 2025.","tagline":"Our frontier-class multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"magistral-small-2506","name":"magistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-0-25-06","open_weight":true,"description":"Our small reasoning model released July 2025.","tagline":"Our small reasoning model released July 2025.","tools":["function_calling"]},{"id":"magistral-small-2507","name":"magistral-small-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-10-31","context_window":40000,"alias":"magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-1-25-07","open_weight":true,"description":"Our small reasoning model released July 2025.","tagline":"Our small reasoning model released July 2025.","tools":["function_calling"]},{"id":"magistral-small-2509","name":"magistral-small-2509","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":128000,"alias":"magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09","open_weight":true,"description":"Our small multimodal reasoning model update of September 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"magistral-small","name":"magistral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":40000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"snapshots":["magistral-small-2509","magistral-small-2507","magistral-small-2506"],"deprecation_date":"2025-10-31","model_type":"reasoning","family":"magistral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/magistral-small-1-0-25-06","open_weight":true,"description":"Our small reasoning model released July 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","tools":["function_calling"]},{"id":"mathstral-7b-01","name":"mathstral-7b-01","created_by":"mistral","source":"official","description":"A math dedicated model for research purposes under an open license.","status":"active","context_window":32000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mathstral-7b-0-1","tagline":"A math dedicated model for research purposes under an open license.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"ministral-14b-2512","name":"ministral-14b-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"alias":"ministral-14b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"model_type":"chat","license":"apache-2.0","parameters":14,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-14b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-14b","name":"ministral-14b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"snapshots":["ministral-14b-2512"],"model_type":"chat","license":"apache-2.0","parameters":14,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-14b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-3b-2410","name":"ministral-3b-2410","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"alias":"ministral-3b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"ministral-3b-2512","name":"ministral-3b-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"alias":"ministral-3b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-3b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-3b","name":"ministral-3b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"snapshots":["ministral-3b-2512","ministral-3b-2410"],"model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"ministral-8b-2410","name":"ministral-8b-2410","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"alias":"ministral-8b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","license":"apache-2.0","parameters":8,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-8b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"ministral-8b-2512","name":"ministral-8b-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"active","context_window":256000,"alias":"ministral-8b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","license":"apache-2.0","parameters":8,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-3-8b-25-12","open_weight":true,"tools":["function_calling"]},{"id":"ministral-8b","name":"ministral-8b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"ministral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"snapshots":["ministral-8b-2512","ministral-8b-2410"],"model_type":"chat","license":"apache-2.0","parameters":8,"page_url":"https://docs.mistral.ai/models/model-cards/ministral-8b-24-1","open_weight":true,"tools":["function_calling"]},{"id":"mistral-7b-01","name":"mistral-7b-01","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":8000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-7b","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/mistral-7b-0-1","open_weight":true,"description":"Our very first open model, SOTA competing against models up to 2 to 3x larger.","tagline":"Our very first open model, SOTA competing against models up to 2 to 3x larger."},{"id":"mistral-7b-02","name":"mistral-7b-02","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-7b","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/mistral-7b-0-2","open_weight":true,"description":"An update to our first open model with better context.","tagline":"An update to our first open model with better context."},{"id":"mistral-7b-03","name":"mistral-7b-03","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-7b","license":"apache-2.0","parameters":7,"page_url":"https://docs.mistral.ai/models/model-cards/mistral-7b-0-3","open_weight":true,"description":"A minor update to our first open model.","tagline":"A minor update to our first open model."},{"id":"mistral-embed-2312","name":"mistral-embed-2312","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":8000,"alias":"mistral-embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"embed","family":"mistral-embed","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-embed-23-12","open_weight":true,"tools":["function_calling"]},{"id":"mistral-embed","name":"mistral-embed","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":8000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["mistral-embed-2312"],"model_type":"embed","family":"mistral-embed","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-embed-23-12","open_weight":true,"tools":["function_calling"]},{"id":"mistral-large-2402","name":"mistral-large-2402","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","deprecation_date":"2024-11-30","context_window":32000,"alias":"mistral-large","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-1-0-24-02","open_weight":true,"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities."},{"id":"mistral-large-2407","name":"mistral-large-2407","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","deprecation_date":"2024-11-30","context_window":128000,"alias":"mistral-large","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-2-0-24-07","open_weight":true,"description":"A powerfull model with open weights under a MRL license, competing agains SOTA much larger models.","tagline":"A powerfull model with open weights under a MRL license, competing agains SOTA much larger models."},{"id":"mistral-large-2411","name":"mistral-large-2411","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","context_window":128000,"alias":"mistral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-2-1-24-11","open_weight":true,"description":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024.","tagline":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"mistral-large-2512","name":"mistral-large-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"active","context_window":256000,"alias":"mistral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-3-25-12","open_weight":true,"tools":["function_calling"],"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities."},{"id":"mistral-large","name":"mistral-large","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-large","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.5},"snapshots":["mistral-large-2512","mistral-large-2411","mistral-large-2407","mistral-large-2402"],"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-large-1-0-24-02","open_weight":true,"tools":["function_calling"],"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024."},{"id":"mistral-medium-2312","name":"mistral-medium-2312","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-medium","status":"deprecated","deprecation_date":"2024-11-30","context_window":32000,"alias":"mistral-medium","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-1-0-23-12","open_weight":false,"description":"Our first SOTA enterprise grade model.","tagline":"Our first SOTA enterprise grade model."},{"id":"mistral-medium-2505","name":"mistral-medium-2505","created_by":"mistral","source":"official","last_updated":"2026-06-18","family":"mistral-medium","status":"deprecated","context_window":128000,"alias":"mistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-25-05","open_weight":false,"description":"Our frontier-class multimodal model released August 2025. Improving tone and performance.","tagline":"Our frontier-class multimodal model released August 2025.","tools":["function_calling"]},{"id":"mistral-medium-2508","name":"mistral-medium-2508","created_by":"mistral","source":"official","last_updated":"2026-06-18","family":"mistral-medium","status":"deprecated","context_window":128000,"alias":"mistral-medium","capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":2},"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-1-25-08","open_weight":false,"description":"Our frontier-class multimodal model released August 2025. Improving tone and performance.","tagline":"Our frontier-class multimodal model released August 2025.","tools":["function_calling"]},{"id":"mistral-medium-3-5-2604","name":"mistral-medium-3-5-2604","created_by":"mistral","source":"official","family":"mistral-medium","description":"Our frontier-class multimodal model optimized for agentic and coding use cases. Released as open weights under a Modified MIT license.","status":"active","context_window":256000,"model_type":"chat","license":"proprietary","alias":"mistral-medium-3-5","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","tagline":"Our frontier-class multimodal model optimized for agentic and coding use cases.","capabilities":{"streaming":true,"tool_call":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"open_weight":false,"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"mistral-medium-3-5","name":"mistral-medium-3-5","created_by":"mistral","source":"official","family":"mistral-medium","description":"Our frontier-class multimodal model optimized for agentic and coding use cases. Released as open weights under a Modified MIT license.","status":"active","context_window":256000,"license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","model_type":"chat","tagline":"Our frontier-class multimodal model optimized for agentic and coding use cases.","capabilities":{"streaming":true,"tool_call":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"open_weight":false,"tools":["function_calling"],"snapshots":["mistral-medium-3-5-2604"],"last_updated":"2026-06-03"},{"id":"mistral-medium-latest","name":"mistral-medium-latest","created_by":"mistral","source":"official","family":"mistral-medium","description":"Our frontier-class multimodal model optimized for agentic and coding use cases. Released as open weights under a Modified MIT license.","status":"active","context_window":256000,"license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-3-5-26-04","model_type":"chat","tagline":"Our frontier-class multimodal model optimized for agentic and coding use cases.","capabilities":{"streaming":true,"tool_call":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"open_weight":false,"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"mistral-medium","name":"mistral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-10","family":"mistral-medium","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2},"snapshots":["mistral-medium-2508","mistral-medium-2505","mistral-medium-2312"],"model_type":"chat","license":"proprietary","page_url":"https://docs.mistral.ai/models/model-cards/mistral-medium-1-0-23-12","open_weight":false,"description":"Our first SOTA enterprise grade model.","tagline":"Our frontier-class multimodal model released August 2025.","tools":["function_calling"]},{"id":"mistral-moderation-2411","name":"mistral-moderation-2411","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":8000,"alias":"mistral-moderation","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"moderation","family":"mistral-moderation","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-moderation-24-11","open_weight":true,"description":"Our latest moderation model with 128k context window and jailbreaking detection. Strong performance on complex multilingual data including long multi-turn conversations.","tagline":"Our latest moderation model with 128k context window and jailbreaking detection, based on Ministral 3B.","tools":["function_calling"],"deprecation_date":"2026-03-31"},{"id":"mistral-moderation-2603","name":"mistral-moderation-2603","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":128000,"alias":"mistral-moderation","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"moderation","family":"mistral-moderation","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-moderation-26-03","open_weight":true,"description":"Our latest moderation model with 128k context window and jailbreaking detection. Strong performance on complex multilingual data including long multi-turn conversations.","tagline":"Our latest moderation model with 128k context window and jailbreaking detection, based on Ministral 3B.","tools":["function_calling"]},{"id":"mistral-moderation","name":"mistral-moderation","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":8000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["mistral-moderation-2603","mistral-moderation-2411"],"model_type":"moderation","family":"mistral-moderation","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-moderation-24-11","open_weight":true,"description":"Our latest moderation model with 128k context window and jailbreaking detection. Strong performance on complex multilingual data including long multi-turn conversations.","tagline":"Our latest moderation model with 128k context window and jailbreaking detection, based on Ministral 3B.","tools":["function_calling"]},{"id":"mistral-nemo-2407","name":"mistral-nemo-2407","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":128000,"alias":"mistral-nemo","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","family":"mistral-nemo","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-nemo-12b-24-07","open_weight":true,"description":"Our best multilingual open source model released July 2024.","tagline":"Our best multilingual open source model released July 2024.","tools":["function_calling"]},{"id":"mistral-nemo","name":"mistral-nemo","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"snapshots":["mistral-nemo-2407"],"model_type":"chat","family":"mistral-nemo","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-nemo-12b-24-07","open_weight":true,"description":"Our best multilingual open source model released July 2024.","tagline":"Our best multilingual open source model released July 2024.","tools":["function_calling"]},{"id":"mistral-next","name":"mistral-next","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-next","open_weight":true,"description":"A prototype model, designed to be brief and concise.","tagline":"A prototype model, designed to be brief and concise."},{"id":"mistral-ocr-2503","name":"mistral-ocr-2503","created_by":"mistral","source":"official","description":"Our first OCR optimized model.","status":"deprecated","model_type":"chat","license":"apache-2.0","alias":"mistral-ocr","page_url":"https://docs.mistral.ai/models/model-cards/ocr-25-03","tagline":"Our first OCR optimized model.","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"mistral-ocr-2505","name":"mistral-ocr-2505","created_by":"mistral","source":"official","status":"deprecated","model_type":"chat","license":"apache-2.0","alias":"mistral-ocr","page_url":"https://docs.mistral.ai/models/model-cards/ocr-2-25-05","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-29","description":"Our OCR service powering our Document AI stack that enables our users to extract interleaved text and images. OCR 4 is available as the newer model. OCR 3 remains available for existing integrations and production workloads.","tagline":"Our first OCR optimized model."},{"id":"mistral-ocr-2512","name":"mistral-ocr-2512","created_by":"mistral","source":"official","status":"active","model_type":"chat","license":"apache-2.0","alias":"mistral-ocr","page_url":"https://docs.mistral.ai/models/model-cards/ocr-3-25-12","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-29","description":"Our OCR service powering our Document AI stack that enables our users to extract interleaved text and images. OCR 4 is available as the newer model. OCR 3 remains available for existing integrations and production workloads.","tagline":"Our first OCR optimized model."},{"id":"mistral-ocr-latest","name":"mistral-ocr-latest","created_by":"mistral","source":"official","description":"Our latest OCR service powering our Document AI stack, with native paragraph-level bounding box extraction and structural block labels.","status":"active","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/ocr-4-0","model_type":"chat","tagline":"Our latest OCR service powering our Document AI stack, with native paragraph-level bounding box extraction and structural block labels.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"mistral-ocr","name":"mistral-ocr","created_by":"mistral","source":"official","status":"deprecated","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/ocr-25-03","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["mistral-ocr-2512","mistral-ocr-2505","mistral-ocr-2503"],"last_updated":"2026-06-03","description":"Our first OCR optimized model.","tagline":"Our first OCR optimized model."},{"id":"mistral-saba-2502","name":"mistral-saba-2502","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-06-10","context_window":32000,"alias":"mistral-saba","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","family":"mistral-saba","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-saba-25-02","open_weight":true,"description":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia.","tagline":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia."},{"id":"mistral-saba","name":"mistral-saba","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","deprecation_date":"2025-06-10","context_window":32000,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"snapshots":["mistral-saba-2502"],"model_type":"chat","family":"mistral-saba","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-saba-25-02","open_weight":true,"description":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia.","tagline":"A finetuned Small model trained on meticulously curated datasets from across the Middle East and South Asia."},{"id":"mistral-small-2402","name":"mistral-small-2402","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2024-11-30","context_window":32000,"alias":"mistral-small","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-1-0-24-02","open_weight":true,"description":"An enterprise grade small model for efficiency.","tagline":"An enterprise grade small model for efficiency."},{"id":"mistral-small-2409","name":"mistral-small-2409","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2025-11-06","context_window":32000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-2-0-24-09","open_weight":true,"description":"An update to our efficient small model with open weights under an MRL license.","tagline":"An update to our efficient small model with open weights under an MRL license.","tools":["function_calling"]},{"id":"mistral-small-2501","name":"mistral-small-2501","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2025-11-06","context_window":32000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-3-0-25-01","open_weight":true,"description":"A small efficient and powerfull 24B open model for personal or commercial use.","tagline":"A small efficient and powerfull 24B open model for personal or commercial use.","tools":["function_calling"]},{"id":"mistral-small-2503","name":"mistral-small-2503","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","deprecation_date":"2025-11-06","context_window":128000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-3-1-25-03","open_weight":true,"description":"A new leader in the small models category with image understanding capabilities, released March 2025.","tagline":"A new leader in the small models category with image understanding capabilities, released March 2025.","tools":["function_calling"]},{"id":"mistral-small-2506","name":"mistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","context_window":128000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-3-2-25-06","open_weight":true,"description":"An update to our previous small model, released June 2025.","tagline":"An update to our previous small model, released June 2025.","tools":["function_calling"]},{"id":"mistral-small-2603","name":"mistral-small-2603","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"active","context_window":256000,"alias":"mistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-4-0-26-03","open_weight":true,"description":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model. 119B parameters with 6.5B active.","tagline":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model.","tools":["function_calling"]},{"id":"mistral-small-creative-2512","name":"mistral-small-creative-2512","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"active","context_window":32000,"alias":"mistral-small-creative","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-creative-25-12","open_weight":true,"description":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction following and conversational agents.","tagline":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction fol","tools":["function_calling"]},{"id":"mistral-small-creative","name":"mistral-small-creative","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"active","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"snapshots":["mistral-small-creative-2512"],"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-creative-25-12","open_weight":true,"description":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction following and conversational agents.","tagline":"An experimental specialized small model trained on meticulously curated data, designed for creative writing, narrative generation, roleplay and character-driven dialog, general-purpose instruction fol","tools":["function_calling"]},{"id":"mistral-small","name":"mistral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"mistral-small","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.6},"snapshots":["mistral-small-2603","mistral-small-2506","mistral-small-2503","mistral-small-2501","mistral-small-2409","mistral-small-2402"],"model_type":"chat","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mistral-small-1-0-24-02","open_weight":true,"description":"An enterprise grade small model for efficiency.","tagline":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model.","tools":["function_calling"]},{"id":"mixtral-8x22b-0-1-03","name":"mixtral-8x22b-0-1-03","created_by":"mistral","source":"official","family":"mixtral","description":"A large open MoE model competing with much larger dense models.","status":"active","context_window":64000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mixtral-8x22b-0-1-0-3","model_type":"chat","tagline":"A large open MoE model competing with much larger dense models.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"mixtral-8x7b-01","name":"mixtral-8x7b-01","created_by":"mistral","source":"official","family":"mixtral","description":"Our first open MoE model, SOTA competing against models up to 2 to 3x larger.","status":"active","context_window":32000,"license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/mixtral-8x7b-0-1","model_type":"chat","tagline":"Our first open MoE model, SOTA competing against models up to 2 to 3x larger.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"open_weight":true,"last_updated":"2026-06-03"},{"id":"pixtral-12b-2409","name":"pixtral-12b-2409","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"alias":"pixtral-12b","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"chat","license":"apache-2.0","parameters":12,"page_url":"https://docs.mistral.ai/models/model-cards/pixtral-12b-24-09","open_weight":true,"description":"A 12B model with image understanding capabilities in addition to text.","tagline":"A 12B model with image understanding capabilities in addition to text.","tools":["function_calling"]},{"id":"pixtral-12b","name":"pixtral-12b","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","deprecation_date":"2025-12-02","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["pixtral-12b-2409"],"model_type":"chat","license":"apache-2.0","parameters":12,"page_url":"https://docs.mistral.ai/models/model-cards/pixtral-12b-24-09","open_weight":true,"description":"A 12B model with image understanding capabilities in addition to text.","tagline":"A 12B model with image understanding capabilities in addition to text.","tools":["function_calling"]},{"id":"pixtral-large-2411","name":"pixtral-large-2411","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","context_window":128000,"alias":"pixtral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":6},"model_type":"chat","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/pixtral-large-24-11","open_weight":true,"tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"pixtral-large","name":"pixtral-large","created_by":"mistral","source":"official","last_updated":"2026-06-03","family":"pixtral","status":"deprecated","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":6},"snapshots":["pixtral-large-2411"],"model_type":"chat","license":"mnpl","page_url":"https://docs.mistral.ai/models/model-cards/pixtral-large-24-11","open_weight":true,"tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"voxtral-mini-2507","name":"voxtral-mini-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":32000,"alias":"voxtral-mini","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-25-07","open_weight":true,"description":"A mini version of our first audio input model.","tagline":"An efficient audio input model, fine-tuned and optimized for transcription purposes only.","tools":["function_calling"],"deprecation_date":"2026-02-27"},{"id":"voxtral-mini-2602","name":"voxtral-mini-2602","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"active","alias":"voxtral-mini","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"context_window":32000,"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-transcribe-26-02","open_weight":true,"description":"An efficient audio input model, pre-trained and optimized for transcription purposes only.","tagline":"An efficient audio input model, fine-tuned and optimized for transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-mini-transcribe-realtime-2602","name":"voxtral-mini-transcribe-realtime-2602","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"active","alias":"voxtral-mini-transcribe-realtime","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"transcription","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-transcribe-realtime-26-02","open_weight":true,"description":"An efficient audio input model, pre-trained and optimized for live transcription purposes only.","tagline":"An efficient audio input model, fine-tuned and optimized for live transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-mini-transcribe-realtime","name":"voxtral-mini-transcribe-realtime","created_by":"mistral","source":"official","last_updated":"2026-06-10","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["voxtral-mini-transcribe-realtime-2602"],"model_type":"transcription","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-transcribe-realtime-26-02","open_weight":true,"description":"An efficient audio input model, pre-trained and optimized for live transcription purposes only.","tagline":"An efficient audio input model, fine-tuned and optimized for live transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-mini-tts-2603","name":"voxtral-mini-tts-2603","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~90ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","model_type":"tts","license":"apache-2.0","alias":"voxtral-mini-tts","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"voxtral-mini-tts","name":"voxtral-mini-tts","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~90ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03","model_type":"tts","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["voxtral-mini-tts-2603"],"last_updated":"2026-06-03"},{"id":"voxtral-mini","name":"voxtral-mini","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"deprecated","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["voxtral-mini-2507","voxtral-mini-2602"],"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-mini-25-07","open_weight":true,"description":"A mini version of our first audio input model.","tagline":"An efficient audio input model, fine-tuned and optimized for transcription purposes only.","tools":["function_calling"]},{"id":"voxtral-small-2507","name":"voxtral-small-2507","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"alias":"voxtral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-small-25-07","open_weight":true,"description":"Our first model with audio input capabilities for instruct use cases.","tagline":"Our first model with audio input capabilities for instruct use cases.","tools":["function_calling"]},{"id":"voxtral-small","name":"voxtral-small","created_by":"mistral","source":"official","last_updated":"2026-06-03","status":"active","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"snapshots":["voxtral-small-2507"],"model_type":"audio","family":"voxtral","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/model-cards/voxtral-small-25-07","open_weight":true,"description":"Our first model with audio input capabilities for instruct use cases.","tagline":"Our first model with audio input capabilities for instruct use cases.","tools":["function_calling"]},{"id":"voxtral-tts-2603","name":"voxtral-tts-2603","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~100ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","model_type":"tts","license":"apache-2.0","alias":"voxtral-tts","page_url":"https://docs.mistral.ai/models/voxtral-tts-26-03","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"last_updated":"2026-03-27"},{"id":"voxtral-tts","name":"voxtral-tts","created_by":"mistral","source":"official","family":"voxtral","description":"Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~100ms time-to-first-audio, and no transcript required for voice prompts.","status":"active","license":"apache-2.0","page_url":"https://docs.mistral.ai/models/voxtral-tts-26-03","model_type":"tts","tagline":"Our state-of-the-art text-to-speech model with zero-shot voice cloning.","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"open_weight":true,"tools":["function_calling"],"snapshots":["voxtral-tts-2603"],"last_updated":"2026-03-27"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/novita.ts b/packages/npm/src/providers/novita.ts
index b0d43a124..94ccbbc99 100644
--- a/packages/npm/src/providers/novita.ts
+++ b/packages/npm/src/providers/novita.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"novita","name":"Novita AI","region":"US","url":"https://novita.ai","api_url":"https://api.novita.ai/v3/openai","docs_url":"https://docs.novita.ai","pricing_url":"https://novita.ai/pricing","description":"Cloud AI inference platform offering OpenAI-compatible access to language, image, video, and audio models.","type":"cloud","models_url":"https://api.novita.ai/v3/openai/models","openai_compatible":true,"free_tier":false,"twitter_url":"https://x.com/NovitaAI","icon":"","models":[{"id":"ai_infer_test_1","name":"ai_infer_test_1","created_by":"novita","source":"official","status":"active","release_date":"2025-11-21","context_window":200000,"max_output_tokens":200000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"ai_infer_test_2","name":"ai_infer_test_2","created_by":"novita","source":"official","status":"active","release_date":"2025-11-21","context_window":200000,"max_output_tokens":200000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"ai_infer_test_3","name":"ai_infer_test_3","created_by":"novita","source":"official","status":"active","release_date":"2025-11-21","context_window":200000,"max_output_tokens":200000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"baichuan/baichuan-m2-32b","name":"BaiChuan M2 32B","created_by":"baichuan","source":"official","description":"Baichuan-M2 is a medically-enhanced reasoning model specifically designed for real-world medical reasoning tasks.","status":"active","release_date":"2025-08-13","context_window":131072,"max_output_tokens":131072,"model_type":"chat","parameters":32,"tagline":"Baichuan-M2 is a medically-enhanced reasoning model specifically designed for real-world medical reasoning tasks.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.07},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"baidu/cobuddy","name":"CoBuddy","created_by":"baidu","source":"official","description":"CoBuddy is a specialized code generation model developed by Baidu, engineered with targeted optimizations for Coding and AI Agent scenarios.","status":"active","release_date":"2026-06-03","context_window":131072,"max_output_tokens":65536,"model_type":"chat","tagline":"CoBuddy is a specialized code generation model developed by Baidu, engineered with targeted optimizations for Coding and AI Agent scenarios.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10","pricing":{"input":0.28,"output":1.13}},{"id":"baidu/ernie-4.5-21B-a3b-thinking","name":"ERNIE-4.5-21B-A3B-Thinking","created_by":"baidu","source":"official","description":"ERNIE-4.5-21B-A3B-Thinking is a text-based Mixture of Experts (MoE) post-training model featuring 21B total parameters with 3B active parameters per token.","status":"preview","release_date":"2025-09-19","context_window":131072,"max_output_tokens":65536,"model_type":"chat","parameters":21,"active_parameters":3,"tagline":"ERNIE-4.5-21B-A3B-Thinking is a text-based Mixture of Experts (MoE) post-training model featuring 21B total parameters with 3B active parameters per token.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.28},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"baidu/ernie-4.5-21B-a3b","name":"ERNIE 4.5 21B A3B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"active","release_date":"2025-06-30","context_window":120000,"max_output_tokens":8000,"model_type":"chat","parameters":21,"active_parameters":3,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"tool_call":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.07,"output":0.28},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"baidu/ernie-4.5-300b-a47b-paddle","name":"ERNIE 4.5 300B A47B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"preview","release_date":"2025-06-30","context_window":123000,"max_output_tokens":12000,"model_type":"chat","parameters":300,"active_parameters":47,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.28,"output":1.1},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"baidu/ernie-4.5-vl-28b-a3b-thinking","name":"ERNIE-4.5-VL-28B-A3B-Thinking","created_by":"baidu","source":"official","description":"Built upon the powerful ERNIE-4.5-VL-28B-A3B architecture, the newly upgraded ERNIE-4.5-VL-28B-A3B-Thinking achieves a remarkable leap forward in multimodal reasoning capabilities.","status":"preview","release_date":"2025-11-26","context_window":131072,"max_output_tokens":65536,"model_type":"chat","parameters":28,"active_parameters":3,"tagline":"Built upon the powerful ERNIE-4.5-VL-28B-A3B architecture, the newly upgraded ERNIE-4.5-VL-28B-A3B-Thinking achieves a remarkable leap forward in multimodal reasoning capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.39,"output":0.39},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"baidu/ernie-4.5-vl-28b-a3b","name":"ERNIE 4.5 VL 28B A3B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"active","release_date":"2025-06-30","context_window":30000,"max_output_tokens":8000,"model_type":"chat","parameters":28,"active_parameters":3,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.14,"output":0.56},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"baidu/ernie-4.5-vl-424b-a47b","name":"ERNIE 4.5 VL 424B A47B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"active","release_date":"2025-06-30","context_window":123000,"max_output_tokens":16000,"model_type":"chat","parameters":424,"active_parameters":47,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.42,"output":1.25},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"bunny","name":"Bunny","created_by":"novita","source":"official","status":"active","release_date":"2026-04-10","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-ocr-2","name":"DeepSeek-OCR 2","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-OCR 2 is a multimodal document recognition model released by DeepSeek AI, serving as an upgrade to DeepSeek-OCR.","status":"active","release_date":"2026-01-27","context_window":8192,"max_output_tokens":8192,"model_type":"chat","tagline":"DeepSeek-OCR 2 is a multimodal document recognition model released by DeepSeek AI, serving as an upgrade to DeepSeek-OCR.","capabilities":{"vision":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-ocr","name":"DeepSeek-OCR","created_by":"deepseek","source":"official","family":"deepseek","status":"active","release_date":"2025-10-24","context_window":8192,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-prover-v2-671b","name":"Deepseek Prover V2 671B","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek Launches Open-Source Model DeepSeek-Prover-V2-671B, Specializing in Mathematical Theorem Proving\nThe new model employs a Mixture of Experts (MoE) architecture and is trained using the Lean 4 ","status":"active","release_date":"2025-04-30","context_window":160000,"max_output_tokens":160000,"model_type":"chat","parameters":671,"tagline":"DeepSeek Launches Open-Source Model DeepSeek-Prover-V2-671B, Specializing in Mathematical Theorem Proving\nThe new model employs a Mixture of Experts (MoE) architecture and is trained using the Lean 4 ","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.5},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-0528-qwen3-8b","name":"DeepSeek R1 0528 Qwen3 8B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek-R1-0528-Qwen3-8B is a high-performance reasoning model based on the Qwen3 8B Base model, enhanced through the integration of DeepSeek-R1-0528's Chain-of-Thought (CoT) optimization.","status":"active","release_date":"2025-05-29","context_window":128000,"max_output_tokens":32000,"model_type":"chat","parameters":8,"tagline":"DeepSeek-R1-0528-Qwen3-8B is a high-performance reasoning model based on the Qwen3 8B Base model, enhanced through the integration of DeepSeek-R1-0528's Chain-of-Thought (CoT) optimization.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.06,"output":0.09},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-0528","name":"DeepSeek R1 0528","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 0528 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathe","status":"active","release_date":"2025-05-28","context_window":163840,"max_output_tokens":32768,"model_type":"chat","tagline":"DeepSeek R1 0528 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathe","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1/community","name":"DeepSeek R1","created_by":"deepseek","source":"official","family":"deepseek-r1","status":"active","release_date":"2025-02-14","context_window":64000,"max_output_tokens":8000,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":4,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-r1-distill-llama-70b","name":"DeepSeek R1 Distill LLama 70B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 Distill LLama 70B","status":"active","release_date":"2025-01-27","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":70,"tagline":"DeepSeek R1 Distill LLama 70B","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":0.8},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-distill-qwen-14b","name":"DeepSeek R1 Distill Qwen 14B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 Distill Qwen 14B is a distilled large language model based on Qwen 2.5 14B, using outputs from DeepSeek R1.","status":"active","release_date":"2025-02-02","context_window":32768,"max_output_tokens":16384,"model_type":"chat","parameters":14,"tagline":"DeepSeek R1 Distill Qwen 14B is a distilled large language model based on Qwen 2.5 14B, using outputs from DeepSeek R1.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-distill-qwen-32b","name":"DeepSeek R1 Distill Qwen 32B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on Qwen 2.5 32B, using outputs from DeepSeek R1.","status":"active","release_date":"2025-02-02","context_window":64000,"max_output_tokens":32000,"model_type":"chat","parameters":32,"tagline":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on Qwen 2.5 32B, using outputs from DeepSeek R1.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.3},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-turbo","name":"DeepSeek R1 (Turbo)\t","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","status":"active","release_date":"2025-03-05","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1","name":"DeepSeek R1","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","status":"active","release_date":"2025-01-26","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":4,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-v3-0324","name":"DeepSeek V3 0324","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","status":"active","release_date":"2025-03-25","context_window":163840,"max_output_tokens":65536,"model_type":"chat","tagline":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":1.12},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3/community","name":"DeepSeek V3","created_by":"deepseek","source":"official","family":"deepseek","status":"active","release_date":"2025-02-14","context_window":64000,"max_output_tokens":8000,"model_type":"chat","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.89,"output":0.89},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-v3-turbo","name":"DeepSeek V3 (Turbo)\t","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","status":"active","release_date":"2025-03-05","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":1.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.1-terminus","name":"Deepseek V3.1 Terminus","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3.1-Terminus preserves all original model capabilities while resolving key user-reported issues, including:\n- Language consistency: Significantly reducing mixed Chinese-English output and el","status":"active","release_date":"2025-09-22","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"DeepSeek-V3.1-Terminus preserves all original model capabilities while resolving key user-reported issues, including:\n- Language consistency: Significantly reducing mixed Chinese-English output and el","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":1},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.1","name":"DeepSeek V3.1","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report.","status":"active","release_date":"2025-08-21","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoi","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":1},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.2-exp","name":"Deepseek V3.2 Exp","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3.2-Exp is an experimental model introducing the groundbreaking DeepSeek Sparse Attention (DSA) mechanism for enhanced long-context processing efficiency.","status":"active","release_date":"2025-09-29","context_window":163840,"max_output_tokens":65536,"model_type":"chat","tagline":"DeepSeek-V3.2-Exp is an experimental model introducing the groundbreaking DeepSeek Sparse Attention (DSA) mechanism for enhanced long-context processing efficiency.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":0.41},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.2","name":"Deepseek V3.2","created_by":"deepseek","source":"official","family":"deepseek","description":"We introduce DeepSeek-V3.2, a next-generation foundation model designed to unify high computational efficiency with state-of-the-art reasoning and agentic performance.","status":"active","release_date":"2025-12-01","context_window":163840,"max_output_tokens":65536,"model_type":"chat","tagline":"We introduce DeepSeek-V3.2, a next-generation foundation model designed to unify high computational efficiency with state-of-the-art reasoning and agentic performance.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.269,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v4-flash","name":"Deepseek V4 Flash","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Flash is a lightweight model meticulously designed by DeepSeek to deliver the ultimate combination of lightning-fast response times and unmatched cost-effectiveness.","status":"active","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":393216,"model_type":"chat","tagline":"DeepSeek-V4-Flash is a lightweight model meticulously designed by DeepSeek to deliver the ultimate combination of lightning-fast response times and unmatched cost-effectiveness.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.14,"output":0.28},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"deepseek/deepseek-v4-pro","name":"Deepseek V4 Pro","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Pro is the next-generation flagship open-source large language model developed by DeepSeek, delivering comprehensive performance that rivals the world's premier closed-source models.","status":"active","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":393216,"model_type":"chat","tagline":"DeepSeek-V4-Pro is the next-generation flagship open-source large language model developed by DeepSeek, delivering comprehensive performance that rivals the world's premier closed-source models.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.6,"output":3.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"mit","open_weight":true},{"id":"deepseek/deepseek_v3","name":"DeepSeek V3","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","status":"active","release_date":"2025-01-15","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.89,"output":0.89},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"dev/glm46","name":"dev/glm46","created_by":"dev","source":"official","family":"glm","status":"active","release_date":"2025-11-20","context_window":256000,"max_output_tokens":256000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"elephant","name":"Ling-2.6-flash","created_by":"novita","source":"official","status":"active","release_date":"2026-04-10","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"google/gemma-3-12b-it","name":"Gemma3 12B","created_by":"google","source":"official","family":"gemma-3","status":"active","release_date":"2025-09-04","context_window":131072,"max_output_tokens":8192,"model_type":"chat","parameters":12,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.1},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemma-3-27b-it","name":"Gemma 3 27B","created_by":"google","source":"official","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","status":"active","release_date":"2025-03-25","context_window":98304,"max_output_tokens":16384,"model_type":"chat","parameters":27,"tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","capabilities":{"vision":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.119,"output":0.2},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 26B A4B is built for developers who need scalable performance without sacrificing core capabilities.Crucially, it retains the massive 256K-token context window of the 31B model, making it highly competitive for long-context RAG and processing extensive, image-rich document datasets.","status":"active","release_date":"2026-04-02","context_window":262144,"max_output_tokens":131072,"model_type":"chat","parameters":26,"active_parameters":4,"tagline":"Gemma 4 26B A4B is built for developers who need scalable performance without sacrificing core capabilities.Crucially, it retains the massive 256K-token context window of the 31B model, making it high","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.13,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 31B is engineered to tackle the most demanding enterprise workloads and complex reasoning tasks.","status":"active","release_date":"2026-04-02","context_window":262144,"max_output_tokens":131072,"model_type":"chat","parameters":31,"tagline":"Gemma 4 31B is engineered to tackle the most demanding enterprise workloads and complex reasoning tasks.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.14,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"gryphe/mythomax-l2-13b","name":"Mythomax L2 13B","created_by":"gryphe","source":"official","description":"The idea behind this merge is that each layer is composed of several tensors, which are in turn responsible for specific functions.","status":"active","release_date":"2024-04-25","context_window":4096,"max_output_tokens":3200,"model_type":"chat","parameters":13,"tagline":"The idea behind this merge is that each layer is composed of several tensors, which are in turn responsible for specific functions.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.09},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"gt-4p","name":"gt-4p","created_by":"novita","source":"official","description":"chatgpt-4o","status":"active","release_date":"2025-03-31","max_output_tokens":131072,"model_type":"chat","tagline":"chatgpt-4o","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"inclusionai/ling-2.6-1t","name":"Ling-2.6-1T","created_by":"inclusionai","source":"official","status":"active","release_date":"2026-04-23","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"inclusionai/ling-2.6-flash","name":"Ling-2.6-flash","created_by":"inclusionai","source":"official","status":"active","release_date":"2026-04-24","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"inclusionai/ring-2.6-1t","name":"Ring-2.6-1T","created_by":"inclusionai","source":"official","status":"active","release_date":"2026-05-08","context_window":262144,"max_output_tokens":65536,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","pricing":{"input":0.3,"output":2.5}},{"id":"kwaipilot/kat-coder-pro","name":"Kat Coder Pro","created_by":"kwaipilot","source":"official","description":"KAT-Coder-Pro V2 by KwaiKAT is a non-reasoning model optimized for agentic coding.","status":"active","release_date":"2026-01-05","context_window":256000,"max_output_tokens":128000,"model_type":"chat","tagline":"KAT-Coder-Pro V2 by KwaiKAT is a non-reasoning model optimized for agentic coding.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"meta-llama/llama-3-70b-instruct","name":"Llama3 70B Instruct","created_by":"meta","source":"official","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","status":"active","release_date":"2024-04-25","context_window":8192,"max_output_tokens":8000,"model_type":"chat","parameters":70,"tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.51,"output":0.74},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"meta-llama/llama-3-8b-instruct","name":"Llama 3 8B Instruct","created_by":"meta","source":"official","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","status":"active","release_date":"2024-04-25","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":8,"tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.04,"output":0.04},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"meta-llama/llama-3.1-8b-instruct","name":"Llama 3.1 8B Instruct","created_by":"meta","source":"official","family":"llama-3.1","description":"Meta's latest class of models, Llama 3.1, launched with a variety of sizes and configurations.","status":"active","release_date":"2024-07-24","context_window":16384,"max_output_tokens":16384,"model_type":"chat","parameters":8,"tagline":"Meta's latest class of models, Llama 3.1, launched with a variety of sizes and configurations.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.02,"output":0.05},"endpoints":["chat_completions"],"last_updated":"2026-05-11","knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta-llama/llama-3.2-1b-instruct","name":"Llama 3.2 1B Instruct\t","created_by":"meta","source":"official","family":"llama-3.2","description":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).","status":"active","release_date":"2024-11-26","context_window":131000,"max_output_tokens":32000,"model_type":"chat","parameters":1,"tagline":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).","capabilities":{"structured_output":true,"streaming":true,"json_mode":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.02,"output":0.02},"endpoints":["chat_completions"],"last_updated":"2026-06-03","knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta-llama/llama-3.2-3b-instruct","name":"Llama 3.2 3B Instruct","created_by":"meta","source":"official","family":"llama-3.2","description":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out)","status":"active","release_date":"2024-11-26","context_window":32768,"max_output_tokens":32000,"model_type":"chat","parameters":3,"tagline":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out)","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.05},"endpoints":["chat_completions"],"last_updated":"2026-05-11","knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta-llama/llama-3.3-70b-instruct","name":"Llama 3.3 70B Instruct","created_by":"meta","source":"official","family":"llama-3.3","description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","status":"active","release_date":"2024-12-07","context_window":131072,"max_output_tokens":120000,"model_type":"chat","parameters":70,"tagline":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.135,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","knowledge_cutoff":"2023-12","license":"llama3.3","open_weight":true},{"id":"meta-llama/llama-4-maverick-17b-128e-instruct-fp8","name":"Llama 4 Maverick Instruct","created_by":"meta","source":"official","family":"llama-4","description":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward pass (400B total).","status":"active","release_date":"2025-04-06","context_window":1048576,"max_output_tokens":8192,"model_type":"chat","parameters":17,"tagline":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forw","capabilities":{"vision":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.27,"output":0.85},"endpoints":["chat_completions"],"last_updated":"2026-05-11","license":"llama4","open_weight":true},{"id":"meta-llama/llama-4-scout-17b-16e-instruct","name":"Llama 4 Scout Instruct","created_by":"meta","source":"official","family":"llama-4","description":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","status":"active","release_date":"2025-04-06","context_window":131072,"max_output_tokens":131072,"model_type":"chat","parameters":17,"tagline":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","capabilities":{"vision":true,"reasoning":true,"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.18,"output":0.59},"endpoints":["chat_completions"],"last_updated":"2026-05-11","license":"llama4","open_weight":true},{"id":"microsoft/wizardlm-2-8x22b","name":"Wizardlm 2 8x22B","created_by":"microsoft","source":"official","description":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","status":"active","release_date":"2024-04-24","context_window":65535,"max_output_tokens":8000,"model_type":"chat","tagline":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.62,"output":0.62},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"minimax/m2-her","name":"M2-her","created_by":"minimax","source":"official","description":"M2-her测试","status":"active","release_date":"2026-01-22","context_window":32000,"model_type":"chat","tagline":"M2-her测试","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2.1","name":"Minimax M2.1","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.1 is a cutting-edge AI model designed to revolutionize how developers build software.","status":"active","release_date":"2025-12-23","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax M2.1 is a cutting-edge AI model designed to revolutionize how developers build software.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5-highspeed","name":"MiniMax M2.5-highspeed","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.5-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","status":"active","release_date":"2026-03-02","context_window":204800,"max_output_tokens":131100,"model_type":"chat","tagline":"MiniMax M2.5-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5","name":"MiniMax M2.5","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","status":"active","release_date":"2026-02-12","context_window":204800,"max_output_tokens":131100,"model_type":"chat","tagline":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.7-highspeed","name":"MiniMax M2.7-highspeed","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.7-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","status":"active","release_date":"2026-04-02","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax M2.7-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":2.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2.7","name":"MiniMax M2.7","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.7 is an all-around evolved, versatile open-source large language model that seamlessly blends hardcore engineering productivity with high-EQ, human-like interaction capabilities.","status":"active","release_date":"2026-03-17","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax M2.7 is an all-around evolved, versatile open-source large language model that seamlessly blends hardcore engineering productivity with high-EQ, human-like interaction capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2","name":"MiniMax-M2","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","status":"active","release_date":"2025-10-27","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m3","name":"MiniMax M3","created_by":"minimax","source":"official","family":"minimax","status":"active","release_date":"2026-06-01","context_window":1000000,"max_output_tokens":131072,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","license":"proprietary","open_weight":false},{"id":"minimaxai/minimax-m1-80k","name":"MiniMax M1","created_by":"minimaxai","source":"official","family":"minimax","description":"MiniMax-M1: The World's First Open-Weight, Large-Scale Hybrid Attention Inference Model\n\nMiniMax-M1 adopts a Mixture of Experts (MoE) architecture and integrates the Flash Attention mechanism. The mod","status":"active","release_date":"2025-06-17","context_window":1000000,"max_output_tokens":40000,"model_type":"chat","tagline":"MiniMax-M1: The World's First Open-Weight, Large-Scale Hybrid Attention Inference Model\n\nMiniMax-M1 adopts a Mixture of Experts (MoE) architecture and integrates the Flash Attention mechanism. The mod","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"mistralai/mistral-nemo","name":"Mistral Nemo","created_by":"mistral","source":"official","family":"mistral-nemo","description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","status":"active","release_date":"2024-07-30","context_window":60288,"max_output_tokens":16000,"model_type":"chat","tagline":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.04,"output":0.17},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"apache-2.0","open_weight":true},{"id":"moonshotai/kimi-k2-0905","name":"Kimi K2 0905","created_by":"moonshot","source":"official","family":"kimi-k2","description":"Kimi K2 0905 is the September update of Kimi K2 0711.","status":"active","release_date":"2025-09-05","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2 0905 is the September update of Kimi K2 0711.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"moonshotai/kimi-k2-instruct","name":"Kimi K2 Instruct","created_by":"moonshot","source":"official","family":"kimi-k2","description":"Kimi K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters.","status":"active","release_date":"2025-07-11","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"Kimi K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.57,"output":2.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"moonshotai/kimi-k2-thinking","name":"Kimi K2 Thinking","created_by":"moonshot","source":"official","family":"kimi-k2","description":"The kimi-k2-thinking model is a general-purpose agentic reasoning model developed by Moonshot AI.","status":"active","release_date":"2025-11-07","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"The kimi-k2-thinking model is a general-purpose agentic reasoning model developed by Moonshot AI.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.5","name":"Kimi K2.5","created_by":"moonshot","source":"official","family":"kimi-k2.5","description":"Kimi K2.5 is the latest flagship iteration of Moonshot AI's large language model series, representing a significant leap in multimodal and agentic capabilities.","status":"active","release_date":"2026-01-27","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.5 is the latest flagship iteration of Moonshot AI's large language model series, representing a significant leap in multimodal and agentic capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.6","name":"Kimi K2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is an open-source, native multimodal agentic model that significantly advances practical capabilities in long-horizon coding, coding-driven design, and swarm-based task orchestration.","status":"active","release_date":"2026-04-20","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.6 is an open-source, native multimodal agentic model that significantly advances practical capabilities in long-horizon coding, coding-driven design, and swarm-based task orchestration.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.8,"output":3.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"moonshotai/kimi-k2.7-code","name":"Kimi K2.7 Code","created_by":"moonshot","source":"official","family":"kimi-k2.7","description":"Kimi K2.7 Code is MoonshotAI's strongest coding & agentic model — a 1T-parameter MoE (32B activated) , 256K context and interleaved thinking with multi-step tool calling.","status":"active","release_date":"2026-06-12","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.7 Code is MoonshotAI's strongest coding & agentic model — a 1T-parameter MoE (32B activated) , 256K context and interleaved thinking with multi-step tool calling.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.95,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"nex-agi/nex-n2-pro","name":"Nex-N2-Pro","created_by":"nex-agi","source":"official","description":"Nex-N2-Pro is Nex AGI's open-source agentic reasoning model, post-trained on Qwen3.5-397B-A17B (MoE).","status":"active","release_date":"2026-06-04","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Nex-N2-Pro is Nex AGI's open-source agentic reasoning model, post-trained on Qwen3.5-397B-A17B (MoE).","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nousresearch/hermes-2-pro-llama-3-8b","name":"Hermes 2 Pro Llama 3 8B","created_by":"nousresearch","source":"official","description":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mod","status":"active","release_date":"2024-06-27","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":8,"tagline":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mod","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.14,"output":0.14},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"nousresearch/nous-hermes-llama2-13b","name":"Nous Hermes Llama2 13B","created_by":"nousresearch","source":"official","description":"Nous-Hermes-Llama2-13b is a state-of-the-art language model fine-tuned on over 300,000 instructions.","status":"preview","release_date":"2024-04-25","context_window":4096,"max_output_tokens":32768,"model_type":"chat","parameters":13,"tagline":"Nous-Hermes-Llama2-13b is a state-of-the-art language model fine-tuned on over 300,000 instructions.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.17,"output":0.17},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"Nemotron 3 Nano 30B A3B","created_by":"nvidia","source":"official","description":"Nemotron-3-Nano-30B-A3B is NVIDIA's compute-efficient, open-weight reasoning model built for agentic AI.","status":"active","release_date":"2026-06-09","context_window":262144,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"Nemotron-3-Nano-30B-A3B is NVIDIA's compute-efficient, open-weight reasoning model built for agentic AI.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"openai/gpt-oss-120b","name":"OpenAI GPT OSS 120B","created_by":"openai","source":"official","family":"gpt-oss","description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","status":"active","release_date":"2025-08-06","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":120,"tagline":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":false,"vision":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.25},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b","name":"OpenAI: GPT OSS 20B","created_by":"openai","source":"official","family":"gpt-oss","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","status":"active","release_date":"2025-08-06","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":20,"tagline":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":false,"vision":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.04,"output":0.15},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openchat/openchat-7b","name":"OpenChat 7B","created_by":"openchat","source":"official","description":"OpenChat 7B is a library of open-source language models, fine-tuned with \"C-RLFT (Conditioned Reinforcement Learning Fine-Tuning)\" - a strategy inspired by offline reinforcement learning.","status":"preview","release_date":"2024-07-26","context_window":4096,"max_output_tokens":4096,"model_type":"chat","parameters":7,"tagline":"OpenChat 7B is a library of open-source language models, fine-tuned with \"C-RLFT (Conditioned Reinforcement Learning Fine-Tuning)\" - a strategy inspired by offline reinforcement learning.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.06,"output":0.06},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"paddlepaddle/paddleocr-vl","name":"PaddleOCR-VL","created_by":"paddlepaddle","source":"official","status":"active","release_date":"2025-10-22","context_window":16384,"max_output_tokens":16384,"model_type":"chat","capabilities":{"vision":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.02,"output":0.02},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen-2-7b-instruct","name":"Qwen 2 7B Instruct","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen2 is the newest series in the Qwen large language model family.","status":"preview","release_date":"2024-09-18","context_window":32768,"max_output_tokens":32768,"model_type":"chat","parameters":7,"tagline":"Qwen2 is the newest series in the Qwen large language model family.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.054,"output":0.054},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"qwen/qwen-2-vl-72b-instruct","name":"Qwen 2 VL 72B Instruct","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen2 VL 72B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\nSoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on ","status":"preview","release_date":"2024-12-19","context_window":32768,"max_output_tokens":120000,"model_type":"chat","parameters":72,"tagline":"Qwen2 VL 72B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\nSoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on ","capabilities":{"vision":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.45,"output":0.45},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"qwen/qwen-2.5-72b-instruct","name":"Qwen 2.5 72B Instruct","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen2.5 is the latest series of Qwen large language models.","status":"active","release_date":"2024-10-15","context_window":32000,"max_output_tokens":8192,"model_type":"chat","parameters":72,"tagline":"Qwen2.5 is the latest series of Qwen large language models.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.38,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen-mt-plus","name":"Qwen MT Plus","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen-MT is a large language model optimized for machine translation, built upon the foundation of the Tongyi Qianwen model.","status":"active","release_date":"2025-09-03","context_window":16384,"max_output_tokens":8192,"model_type":"chat","tagline":"Qwen-MT is a large language model optimized for machine translation, built upon the foundation of the Tongyi Qianwen model.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.25,"output":0.75},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen2.5-7b-instruct","name":"Qwen2.5 7B Instruct","created_by":"alibaba","source":"official","family":"qwen2.5","description":"Qwen2.5 is the latest series of Qwen large language models.","status":"preview","release_date":"2025-04-16","context_window":32000,"max_output_tokens":8192,"model_type":"chat","parameters":7,"tagline":"Qwen2.5 is the latest series of Qwen large language models.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.07},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen2.5-vl-72b-instruct","name":"Qwen2.5 VL 72B Instruct","created_by":"alibaba","source":"official","family":"qwen2.5","description":"Qwen2.5-VL, the latest vision-language model in the Qwen2.5 series, delivers enhanced multimodal capabilities including advanced visual comprehension for object/text recognition, chart/layout analysis, and agent-based dynamic tool orchestration.","status":"preview","release_date":"2025-03-25","context_window":32768,"max_output_tokens":32768,"model_type":"chat","parameters":72,"tagline":"Qwen2.5-VL, the latest vision-language model in the Qwen2.5 series, delivers enhanced multimodal capabilities including advanced visual comprehension for object/text recognition, chart/layout analysis","capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.8,"output":0.8},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3-235b-a22b-fp8","name":"Qwen3 235B A22B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of inference and non-inference modes, enabling seamless switching between modes during conversations.","status":"active","release_date":"2025-04-29","context_window":40960,"max_output_tokens":20000,"model_type":"chat","parameters":235,"active_parameters":22,"tagline":"Achieves effective integration of inference and non-inference modes, enabling seamless switching between modes during conversations.","capabilities":{"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.2,"output":0.8},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","status":"active","release_date":"2025-07-22","context_window":131072,"max_output_tokens":16384,"model_type":"chat","parameters":235,"active_parameters":22,"tagline":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.58},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-235b-a22b-thinking-2507","name":"Qwen3 235B A22b Thinking 2507","created_by":"alibaba","source":"official","family":"qwen3","description":"The Qwen3-235B-A22B-Thinking-2507 represents the newest thinking-enabled model in the Qwen3 series, delivering groundbreaking improvements in reasoning capabilities.","status":"active","release_date":"2025-07-25","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":235,"active_parameters":22,"tagline":"The Qwen3-235B-A22B-Thinking-2507 represents the newest thinking-enabled model in the Qwen3 series, delivering groundbreaking improvements in reasoning capabilities.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-30b-a3b-fp8","name":"Qwen3 30B A3B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","status":"active","release_date":"2025-04-29","context_window":40960,"max_output_tokens":20000,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.45},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-32b-fp8","name":"Qwen3 32B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","status":"active","release_date":"2025-04-29","context_window":40960,"max_output_tokens":20000,"model_type":"chat","parameters":32,"tagline":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.45},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-4b-fp8","name":"Qwen3 4B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless switching during conversations.","status":"preview","release_date":"2025-04-29","context_window":128000,"max_output_tokens":8192,"model_type":"chat","parameters":4,"tagline":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless switching during conversations.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3-8b-fp8","name":"Qwen3 8B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless mode switching during conversations.","status":"active","release_date":"2025-04-29","context_window":128000,"max_output_tokens":20000,"model_type":"chat","parameters":8,"tagline":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless mode switching during conversations.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.035,"output":0.138},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-coder-30b-a3b-instruct","name":"Qwen3 Coder 30b A3B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.","status":"active","release_date":"2025-10-09","context_window":160000,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, an","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.27},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-coder-480b-a35b-instruct","name":"Qwen3 Coder 480B A35B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Coder-480B-A35B-Instruct is a cutting-edge open coding model from Qwen, matching Claude Sonnet’s performance in agentic programming, browser automation, and core development tasks.","status":"active","release_date":"2025-07-23","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":480,"active_parameters":35,"tagline":"Qwen3-Coder-480B-A35B-Instruct is a cutting-edge open coding model from Qwen, matching Claude Sonnet’s performance in agentic programming, browser automation, and core development tasks.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.38,"output":1.55},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3-coder-next","name":"Qwen3 Coder Next","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Coder-Next is an open-weight language model specifically engineered for coding agents and local development environments.","status":"active","release_date":"2026-02-03","context_window":262144,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3-Coder-Next is an open-weight language model specifically engineered for coding agents and local development environments.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.2,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-max","name":"Qwen3 Max","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen/qwen3-max, Enhanced with specialized upgrades in agent programming and tool calling.","status":"active","release_date":"2025-09-24","context_window":262144,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen/qwen3-max, Enhanced with specialized upgrades in agent programming and tool calling.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.11,"output":8.45},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"Qwen3 Next 80B A3B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","status":"active","release_date":"2025-09-10","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":80,"active_parameters":3,"tagline":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"Qwen3 Next 80B A3B Thinking","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","status":"active","release_date":"2025-09-10","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":80,"active_parameters":3,"tagline":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-omni-30b-a3b-instruct","name":"Qwen3 Omni 30B A3B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","status":"active","release_date":"2025-09-24","context_window":65536,"max_output_tokens":16384,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","pricing":{"input":0.25,"output":0.97}},{"id":"qwen/qwen3-omni-30b-a3b-thinking","name":"Qwen3 Omni 30B A3B Thinking","created_by":"alibaba","source":"official","family":"qwen3","description":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","status":"active","release_date":"2025-09-24","context_window":65536,"max_output_tokens":16384,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","pricing":{"input":0.25,"output":0.97}},{"id":"qwen/qwen3-vl-235b-a22b-instruct","name":"Qwen3 VL 235B A22B Instruct","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-09-24","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":235,"active_parameters":22,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-235b-a22b-thinking","name":"Qwen3 VL 235B A22B Thinking","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-09-24","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":235,"active_parameters":22,"capabilities":{"vision":true,"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.98,"output":3.95},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-30b-a3b-instruct","name":"qwen/qwen3-vl-30b-a3b-instruct","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-10-11","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.7},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-30b-a3b-thinking","name":"qwen/qwen3-vl-30b-a3b-thinking","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-10-11","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":1},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-8b-instruct","name":"qwen/qwen3-vl-8b-instruct","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-10-17","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":8,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.08,"output":0.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-122b-a10b","name":"Qwen3.5-122B-A10B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5-122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","status":"active","release_date":"2026-02-26","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":122,"active_parameters":10,"tagline":"The Qwen3.5-122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference eff","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":3.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-27b","name":"Qwen3.5-27B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","status":"active","release_date":"2026-02-26","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":27,"tagline":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-35b-a3b","name":"Qwen3.5-35B-A3B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.","status":"active","release_date":"2026-02-26","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":35,"active_parameters":3,"tagline":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inf","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen3.5-397B-A17B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 series 397B-A17B native vision-language model is based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher inference efficiency.","status":"active","release_date":"2026-02-16","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":397,"active_parameters":17,"tagline":"The Qwen3.5 series 397B-A17B native vision-language model is based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher in","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":3.6},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-plus","name":"Qwen3.5-Plus","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 native vision-language series Plus models are based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher inference efficiency.","status":"active","release_date":"2026-02-16","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"The Qwen3.5 native vision-language series Plus models are based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher infer","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.6-27b","name":"Qwen3.6-27B","created_by":"alibaba","source":"official","family":"qwen3.6","description":"The Qwen3.6 27B native vision-language dense model builds upon the 3.5-27B version, with key improvements in agentic coding capabilities and enhanced STEM reasoning and inference skills.","status":"active","release_date":"2026-04-23","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":27,"tagline":"The Qwen3.6 27B native vision-language dense model builds upon the 3.5-27B version, with key improvements in agentic coding capabilities and enhanced STEM reasoning and inference skills.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":3.6},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen3.6-35B-A3B","created_by":"alibaba","source":"official","family":"qwen3.6","description":"The Qwen3.6 35B-A3B native vision-language model is built on a hybrid architecture that integrates linear attention mechanisms with a sparse mixture-of-experts framework, achieving higher inference efficiency.","status":"active","release_date":"2026-04-21","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":35,"active_parameters":3,"tagline":"The Qwen3.6 35B-A3B native vision-language model is built on a hybrid architecture that integrates linear attention mechanisms with a sparse mixture-of-experts framework, achieving higher inference ef","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.248,"output":1.485},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.6-plus","name":"Qwen3.6-Plus","created_by":"alibaba","source":"official","family":"qwen3.6","description":"The Qwen3.6 native vision-language Plus series models demonstrate exceptional performance on par with the current state-of-the-art models, with a significant improvement in overall results compared to the 3.5 series.","status":"active","release_date":"2026-04-21","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"The Qwen3.6 native vision-language Plus series models demonstrate exceptional performance on par with the current state-of-the-art models, with a significant improvement in overall results compared to","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.7-max","name":"Qwen3.7-Max","created_by":"alibaba","source":"official","family":"qwen3.7","description":"The Qwen3.7 Max model—the largest and most powerful in the series—is currently available for users to try out its text-only features.","status":"active","release_date":"2026-05-21","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"The Qwen3.7 Max model—the largest and most powerful in the series—is currently available for users to try out its text-only features.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.25,"output":3.75},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"sao10k/l3-70b-euryale-v2.1","name":"L3 70B Euryale V2.1\t","created_by":"sao10k","source":"official","description":"The uncensored llama3 model is a powerhouse of creativity, excelling in both roleplay and story writing.","status":"active","release_date":"2024-06-18","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":70,"tagline":"The uncensored llama3 model is a powerhouse of creativity, excelling in both roleplay and story writing.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.48,"output":1.48},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"sao10k/l3-8b-lunaris","name":"Sao10k L3 8B Lunaris\t","created_by":"sao10k","source":"official","description":"A generalist / roleplaying model merge based on Llama 3.","status":"active","release_date":"2024-11-28","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":8,"tagline":"A generalist / roleplaying model merge based on Llama 3.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.05},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"Sao10K/L3-8B-Stheno-v3.2","name":"L3 8B Stheno V3.2","created_by":"sao10k","source":"official","description":"Sao10K/L3-8B-Stheno-v3.2 is a highly skilled actor that excels at fully immersing itself in any role assigned.","status":"active","release_date":"2024-11-29","context_window":8192,"max_output_tokens":32000,"model_type":"chat","parameters":8,"tagline":"Sao10K/L3-8B-Stheno-v3.2 is a highly skilled actor that excels at fully immersing itself in any role assigned.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.05},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"sao10k/l31-70b-euryale-v2.2","name":"L31 70B Euryale V2.2","created_by":"sao10k","source":"official","description":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from Sao10k.","status":"active","release_date":"2024-09-19","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":70,"tagline":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from Sao10k.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.48,"output":1.48},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"stepfun/step-3.7-flash","name":"Step 3.7 Flash","created_by":"stepfun","source":"official","description":"Step-3.7-Flash is StepFun's flagship high-efficiency multimodal reasoning model, built on a sparse Mixture-of-Experts architecture (198B total / ~11B active parameters) that delivers high throughput and low latency for real-time agentic workflows and high-frequency calls.","status":"active","release_date":"2026-06-12","context_window":262144,"max_output_tokens":256000,"model_type":"chat","tagline":"Step-3.7-Flash is StepFun's flagship high-efficiency multimodal reasoning model, built on a sparse Mixture-of-Experts architecture (198B total / ~11B active parameters) that delivers high throughput a","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":1.15},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"teknium/openhermes-2.5-mistral-7b","name":"Openhermes2.5 Mistral 7B","created_by":"teknium","source":"official","description":"OpenHermes 2.5 Mistral 7B is a state of the art Mistral Fine-tune, a continuation of OpenHermes 2 model, which trained on additional code datasets.","status":"preview","release_date":"2024-04-24","context_window":4096,"max_output_tokens":8000,"model_type":"chat","parameters":7,"tagline":"OpenHermes 2.5 Mistral 7B is a state of the art Mistral Fine-tune, a continuation of OpenHermes 2 model, which trained on additional code datasets.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.17,"output":0.17},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"thudm/glm-4-32b-0414","name":"GLM-4-32B-0414","created_by":"thudm","source":"official","family":"glm-4","description":"GLM-4-32B-0414 is the latest open-source model in the GLM series, featuring 32 billion parameters.","status":"active","release_date":"2025-04-16","context_window":32000,"max_output_tokens":32000,"model_type":"chat","parameters":32,"tagline":"GLM-4-32B-0414 is the latest open-source model in the GLM series, featuring 32 billion parameters.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":1.66},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"xiaomimimo/mimo-v2-flash","name":"XiaomiMiMo/MiMo-V2-Flash","created_by":"xiaomi","source":"official","family":"mimo","description":"Xiaomi MiMo-V2-Flash is a proprietary MoE model developed by Xiaomi, designed for extreme inference efficiency with 309B total parameters (15B active).","status":"preview","release_date":"2025-12-19","context_window":262144,"max_output_tokens":32000,"model_type":"chat","tagline":"Xiaomi MiMo-V2-Flash is a proprietary MoE model developed by Xiaomi, designed for extreme inference efficiency with 309B total parameters (15B active).","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.11,"output":0.33},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10","knowledge_cutoff":"2024-12","license":"mit","parameters":309,"active_parameters":15,"architecture":"moe","open_weight":true},{"id":"xiaomimimo/mimo-v2-pro","name":"XiaomiMiMo/MiMo-V2-Pro","created_by":"xiaomi","source":"official","family":"mimo","description":"Xiaomi’s MiMo-V2-Pro is a flagship, >1T parameter foundation model with a 1M context window, specifically optimized for advanced agentic workflows.","status":"preview","release_date":"2026-03-20","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"Xiaomi’s MiMo-V2-Pro is a flagship, >1T parameter foundation model with a 1M context window, specifically optimized for advanced agentic workflows.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2,"output":6},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"proprietary","parameters":1000,"active_parameters":42,"architecture":"moe","open_weight":false},{"id":"xiaomimimo/mimo-v2.5-pro","name":"XiaomiMiMo/MiMo-V2.5-Pro","created_by":"xiaomi","source":"official","family":"mimo","description":"MiMo-V2.5-Pro is purpose-built to push the boundaries of complex software engineering and extreme long-horizon tasks.","status":"active","release_date":"2026-05-06","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"MiMo-V2.5-Pro is purpose-built to push the boundaries of complex software engineering and extreme long-horizon tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.522,"output":1.044},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"xiaomimimo/mimo-v2.5","name":"XiaomiMiMo/MiMo-V2.5","created_by":"xiaomi","source":"official","family":"mimo","description":"Xiaomi's native omni-modal model (text / image / video / audio understanding) with Pro-level agentic performance at ~half the inference cost and a 1M-token context — built for cost-efficient, percepti","status":"active","release_date":"2026-05-29","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"mimo-v2.5, a native omni-modal model supporting text, image, video, and audio understanding, equipped with powerful agentic capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.168,"output":0.336},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"zai-org/autoglm-phone-9b-multilingual","name":"AutoGLM-Phone-9B-Multilingual","created_by":"zai","source":"official","description":"Phone Agent is a mobile intelligent assistant framework built on AutoGLM, capable of understanding smartphone screens through multimodal perception and executing automated operations to complete tasks.","status":"active","release_date":"2025-12-10","context_window":65536,"max_output_tokens":65536,"model_type":"chat","parameters":9,"tagline":"Phone Agent is a mobile intelligent assistant framework built on AutoGLM, capable of understanding smartphone screens through multimodal perception and executing automated operations to complete tasks","capabilities":{"vision":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.035,"output":0.138},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"zai-org/glm-4.5-air","name":"zai-org/glm-4.5-air","created_by":"zai","source":"official","family":"glm-4.5","status":"active","release_date":"2025-10-13","context_window":131072,"max_output_tokens":98304,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.13,"output":0.85},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"zai-org/glm-4.5","name":"GLM-4.5","created_by":"zai","source":"official","family":"glm-4.5","description":"GLM-4.5 Series Models are foundation models specifically engineered for intelligent agents.","status":"active","release_date":"2025-07-28","context_window":131072,"max_output_tokens":98304,"model_type":"chat","tagline":"GLM-4.5 Series Models are foundation models specifically engineered for intelligent agents.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true,"structured_output":true,"prompt_caching":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","parameters":355,"active_parameters":32,"architecture":"moe","open_weight":true},{"id":"zai-org/glm-4.5v","name":"GLM 4.5V","created_by":"zai","source":"official","family":"glm-4.5","description":"Z.ai's GLM-4.5V sets a new standard in visual reasoning, achieving SOTA performance across 42 benchmarks among open-source models.","status":"active","release_date":"2025-08-11","context_window":65536,"max_output_tokens":16384,"model_type":"chat","tagline":"Z.ai's GLM-4.5V sets a new standard in visual reasoning, achieving SOTA performance across 42 benchmarks among open-source models.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":1.8},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","parameters":106,"active_parameters":12,"architecture":"moe","open_weight":true},{"id":"zai-org/glm-4.6","name":"GLM 4.6","created_by":"zai","source":"official","family":"glm-4.6","description":"As the latest iteration in the GLM series, GLM-4.6 achieves comprehensive enhancements across multiple domains, including real-world coding, long-context processing, reasoning, searching, writing, and agentic applications.","status":"active","release_date":"2025-09-30","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"As the latest iteration in the GLM series, GLM-4.6 achieves comprehensive enhancements across multiple domains, including real-world coding, long-context processing, reasoning, searching, writing, and","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-4.6v","name":"GLM 4.6V","created_by":"zai","source":"official","family":"glm-4.6","description":"GLM-4.6V represents a significant multimodal advancement in the GLM series, featuring a 128k-token training context window and achieving state-of-the-art visual understanding accuracy for models of its parameter scale.","status":"active","release_date":"2025-12-08","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"GLM-4.6V represents a significant multimodal advancement in the GLM series, featuring a 128k-token training context window and achieving state-of-the-art visual understanding accuracy for models of it","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":0.9},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-4.7-flash","name":"GLM-4.7-Flash","created_by":"zai","source":"official","family":"glm-4.7","description":"GLM-4.7-Flash, a state-of-the-art model in the 30B class, delivers a compelling balance of high performance and efficiency.","status":"active","release_date":"2026-01-19","context_window":200000,"max_output_tokens":128000,"model_type":"chat","tagline":"GLM-4.7-Flash, a state-of-the-art model in the 30B class, delivers a compelling balance of high performance and efficiency.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-4.7-h","name":"GLM-4.7","created_by":"zai","source":"official","family":"glm-4.7","description":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","status":"active","release_date":"2026-01-26","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"zai-org/glm-4.7","name":"GLM-4.7","created_by":"zai","source":"official","family":"glm-4.7","description":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","status":"active","release_date":"2025-12-22","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-5-turbo","name":"GLM-5-Turbo","created_by":"zai","source":"official","family":"glm-5","description":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","status":"active","release_date":"2026-03-26","context_window":202800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.2,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false},{"id":"zai-org/glm-5.1","name":"GLM-5.1","created_by":"zai","source":"official","family":"glm-5.1","description":"GLM-5.1 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","status":"active","release_date":"2026-04-07","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5.1 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.38,"output":4.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"mit","open_weight":true},{"id":"zai-org/glm-5.2","name":"GLM 5.2","created_by":"zai","source":"official","family":"glm-5.2","description":"GLM-5.2 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","status":"active","release_date":"2026-06-12","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5.2 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.4,"output":4.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","license":"mit","open_weight":true},{"id":"zai-org/glm-5","name":"GLM-5","created_by":"zai","source":"official","family":"glm-5","description":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","status":"active","release_date":"2026-02-08","context_window":202800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1,"output":3.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","parameters":355,"active_parameters":32,"open_weight":true},{"id":"zai-org/glm-5v-turbo","name":"GLM-5V-Turbo","created_by":"zai","source":"official","family":"glm-5","description":"GLM-5V-Turbo is Z.AI’s first multimodal coding foundation model, built for vision-based coding tasks.","status":"active","release_date":"2026-04-02","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5V-Turbo is Z.AI’s first multimodal coding foundation model, built for vision-based coding tasks.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.2,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"mit","open_weight":true},{"id":"zai-org/glm-ocr","name":"GLM-OCR","created_by":"zai","source":"official","family":"glm","description":"GLM-OCR is a lightweight professional OCR model with only 0.9B parameters, achieving SOTA performance with a score of 94.62 on OmniDocBench V1.5.","status":"active","release_date":"2026-02-04","context_window":32000,"max_output_tokens":32000,"model_type":"chat","tagline":"GLM-OCR is a lightweight professional OCR model with only 0.9B parameters, achieving SOTA performance with a score of 94.62 on OmniDocBench V1.5.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"last_updated":"2026-06-03","license":"proprietary","parameters":0.9,"open_weight":false}]};
+export const provider: ProviderWithModels = {"id":"novita","name":"Novita AI","region":"US","url":"https://novita.ai","api_url":"https://api.novita.ai/v3/openai","docs_url":"https://docs.novita.ai","pricing_url":"https://novita.ai/pricing","description":"Cloud AI inference platform offering OpenAI-compatible access to language, image, video, and audio models.","type":"cloud","models_url":"https://api.novita.ai/v3/openai/models","openai_compatible":true,"free_tier":false,"twitter_url":"https://x.com/NovitaAI","icon":"","models":[{"id":"ai_infer_test_1","name":"ai_infer_test_1","created_by":"novita","source":"official","status":"active","release_date":"2025-11-21","context_window":200000,"max_output_tokens":200000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"ai_infer_test_2","name":"ai_infer_test_2","created_by":"novita","source":"official","status":"active","release_date":"2025-11-21","context_window":200000,"max_output_tokens":200000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"ai_infer_test_3","name":"ai_infer_test_3","created_by":"novita","source":"official","status":"active","release_date":"2025-11-21","context_window":200000,"max_output_tokens":200000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"baichuan/baichuan-m2-32b","name":"BaiChuan M2 32B","created_by":"baichuan","source":"official","description":"Baichuan-M2 is a medically-enhanced reasoning model specifically designed for real-world medical reasoning tasks.","status":"active","release_date":"2025-08-13","context_window":131072,"max_output_tokens":131072,"model_type":"chat","parameters":32,"tagline":"Baichuan-M2 is a medically-enhanced reasoning model specifically designed for real-world medical reasoning tasks.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.07},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"baidu/cobuddy","name":"CoBuddy","created_by":"baidu","source":"official","description":"CoBuddy is a specialized code generation model developed by Baidu, engineered with targeted optimizations for Coding and AI Agent scenarios.","status":"active","release_date":"2026-06-03","context_window":131072,"max_output_tokens":65536,"model_type":"chat","tagline":"CoBuddy is a specialized code generation model developed by Baidu, engineered with targeted optimizations for Coding and AI Agent scenarios.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10","pricing":{"input":0.28,"output":1.13}},{"id":"baidu/ernie-4.5-21B-a3b-thinking","name":"ERNIE-4.5-21B-A3B-Thinking","created_by":"baidu","source":"official","description":"ERNIE-4.5-21B-A3B-Thinking is a text-based Mixture of Experts (MoE) post-training model featuring 21B total parameters with 3B active parameters per token.","status":"preview","release_date":"2025-09-19","context_window":131072,"max_output_tokens":65536,"model_type":"chat","parameters":21,"active_parameters":3,"tagline":"ERNIE-4.5-21B-A3B-Thinking is a text-based Mixture of Experts (MoE) post-training model featuring 21B total parameters with 3B active parameters per token.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.28},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"baidu/ernie-4.5-21B-a3b","name":"ERNIE 4.5 21B A3B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"active","release_date":"2025-06-30","context_window":120000,"max_output_tokens":8000,"model_type":"chat","parameters":21,"active_parameters":3,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"tool_call":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.07,"output":0.28},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"baidu/ernie-4.5-300b-a47b-paddle","name":"ERNIE 4.5 300B A47B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"preview","release_date":"2025-06-30","context_window":123000,"max_output_tokens":12000,"model_type":"chat","parameters":300,"active_parameters":47,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.28,"output":1.1},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"baidu/ernie-4.5-vl-28b-a3b-thinking","name":"ERNIE-4.5-VL-28B-A3B-Thinking","created_by":"baidu","source":"official","description":"Built upon the powerful ERNIE-4.5-VL-28B-A3B architecture, the newly upgraded ERNIE-4.5-VL-28B-A3B-Thinking achieves a remarkable leap forward in multimodal reasoning capabilities.","status":"preview","release_date":"2025-11-26","context_window":131072,"max_output_tokens":65536,"model_type":"chat","parameters":28,"active_parameters":3,"tagline":"Built upon the powerful ERNIE-4.5-VL-28B-A3B architecture, the newly upgraded ERNIE-4.5-VL-28B-A3B-Thinking achieves a remarkable leap forward in multimodal reasoning capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.39,"output":0.39},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"baidu/ernie-4.5-vl-28b-a3b","name":"ERNIE 4.5 VL 28B A3B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"active","release_date":"2025-06-30","context_window":30000,"max_output_tokens":8000,"model_type":"chat","parameters":28,"active_parameters":3,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.14,"output":0.56},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"baidu/ernie-4.5-vl-424b-a47b","name":"ERNIE 4.5 VL 424B A47B","created_by":"baidu","source":"official","description":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","status":"active","release_date":"2025-06-30","context_window":123000,"max_output_tokens":16000,"model_type":"chat","parameters":424,"active_parameters":47,"tagline":"The ERNIE 4.5 series of open-source models adopts a Mixture-of-Experts (MoE) architecture, representing an innovative multimodal heterogeneous model structure.","capabilities":{"vision":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.42,"output":1.25},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"bunny","name":"Bunny","created_by":"novita","source":"official","status":"active","release_date":"2026-04-10","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-ocr-2","name":"DeepSeek-OCR 2","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-OCR 2 is a multimodal document recognition model released by DeepSeek AI, serving as an upgrade to DeepSeek-OCR.","status":"active","release_date":"2026-01-27","context_window":8192,"max_output_tokens":8192,"model_type":"chat","tagline":"DeepSeek-OCR 2 is a multimodal document recognition model released by DeepSeek AI, serving as an upgrade to DeepSeek-OCR.","capabilities":{"vision":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-ocr","name":"DeepSeek-OCR","created_by":"deepseek","source":"official","family":"deepseek","status":"active","release_date":"2025-10-24","context_window":8192,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-prover-v2-671b","name":"Deepseek Prover V2 671B","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek Launches Open-Source Model DeepSeek-Prover-V2-671B, Specializing in Mathematical Theorem Proving\nThe new model employs a Mixture of Experts (MoE) architecture and is trained using the Lean 4 ","status":"preview","release_date":"2025-04-30","context_window":160000,"max_output_tokens":160000,"model_type":"chat","parameters":671,"tagline":"DeepSeek Launches Open-Source Model DeepSeek-Prover-V2-671B, Specializing in Mathematical Theorem Proving\nThe new model employs a Mixture of Experts (MoE) architecture and is trained using the Lean 4 ","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.5},"endpoints":["chat_completions"],"last_updated":"2026-06-29"},{"id":"deepseek/deepseek-r1-0528-qwen3-8b","name":"DeepSeek R1 0528 Qwen3 8B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek-R1-0528-Qwen3-8B is a high-performance reasoning model based on the Qwen3 8B Base model, enhanced through the integration of DeepSeek-R1-0528's Chain-of-Thought (CoT) optimization.","status":"active","release_date":"2025-05-29","context_window":128000,"max_output_tokens":32000,"model_type":"chat","parameters":8,"tagline":"DeepSeek-R1-0528-Qwen3-8B is a high-performance reasoning model based on the Qwen3 8B Base model, enhanced through the integration of DeepSeek-R1-0528's Chain-of-Thought (CoT) optimization.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.06,"output":0.09},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-0528","name":"DeepSeek R1 0528","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 0528 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathe","status":"active","release_date":"2025-05-28","context_window":163840,"max_output_tokens":32768,"model_type":"chat","tagline":"DeepSeek R1 0528 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathe","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1/community","name":"DeepSeek R1","created_by":"deepseek","source":"official","family":"deepseek-r1","status":"active","release_date":"2025-02-14","context_window":64000,"max_output_tokens":8000,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":4,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-r1-distill-llama-70b","name":"DeepSeek R1 Distill LLama 70B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 Distill LLama 70B","status":"active","release_date":"2025-01-27","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":70,"tagline":"DeepSeek R1 Distill LLama 70B","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":0.8},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-distill-qwen-14b","name":"DeepSeek R1 Distill Qwen 14B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 Distill Qwen 14B is a distilled large language model based on Qwen 2.5 14B, using outputs from DeepSeek R1.","status":"active","release_date":"2025-02-02","context_window":32768,"max_output_tokens":16384,"model_type":"chat","parameters":14,"tagline":"DeepSeek R1 Distill Qwen 14B is a distilled large language model based on Qwen 2.5 14B, using outputs from DeepSeek R1.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.15},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-distill-qwen-32b","name":"DeepSeek R1 Distill Qwen 32B","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on Qwen 2.5 32B, using outputs from DeepSeek R1.","status":"active","release_date":"2025-02-02","context_window":64000,"max_output_tokens":32000,"model_type":"chat","parameters":32,"tagline":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on Qwen 2.5 32B, using outputs from DeepSeek R1.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.3},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1-turbo","name":"DeepSeek R1 (Turbo)\t","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","status":"active","release_date":"2025-03-05","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-r1","name":"DeepSeek R1","created_by":"deepseek","source":"official","family":"deepseek-r1","description":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","status":"active","release_date":"2025-01-26","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek R1 is the latest open-source model released by the DeepSeek team, featuring impressive reasoning capabilities, particularly achieving performance comparable to OpenAI's o1 model in mathematic","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":4,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-v3-0324","name":"DeepSeek V3 0324","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","status":"active","release_date":"2025-03-25","context_window":163840,"max_output_tokens":65536,"model_type":"chat","tagline":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":1.12},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3/community","name":"DeepSeek V3","created_by":"deepseek","source":"official","family":"deepseek","status":"active","release_date":"2025-02-14","context_window":64000,"max_output_tokens":8000,"model_type":"chat","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.89,"output":0.89},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"deepseek/deepseek-v3-turbo","name":"DeepSeek V3 (Turbo)\t","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","status":"active","release_date":"2025-03-05","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":1.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.1-terminus","name":"Deepseek V3.1 Terminus","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3.1-Terminus preserves all original model capabilities while resolving key user-reported issues, including:\n- Language consistency: Significantly reducing mixed Chinese-English output and el","status":"active","release_date":"2025-09-22","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"DeepSeek-V3.1-Terminus preserves all original model capabilities while resolving key user-reported issues, including:\n- Language consistency: Significantly reducing mixed Chinese-English output and el","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":1},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.1","name":"DeepSeek V3.1","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report.","status":"active","release_date":"2025-08-21","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoi","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":1},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.2-exp","name":"Deepseek V3.2 Exp","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3.2-Exp is an experimental model introducing the groundbreaking DeepSeek Sparse Attention (DSA) mechanism for enhanced long-context processing efficiency.","status":"active","release_date":"2025-09-29","context_window":163840,"max_output_tokens":65536,"model_type":"chat","tagline":"DeepSeek-V3.2-Exp is an experimental model introducing the groundbreaking DeepSeek Sparse Attention (DSA) mechanism for enhanced long-context processing efficiency.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.27,"output":0.41},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v3.2","name":"Deepseek V3.2","created_by":"deepseek","source":"official","family":"deepseek","description":"We introduce DeepSeek-V3.2, a next-generation foundation model designed to unify high computational efficiency with state-of-the-art reasoning and agentic performance.","status":"active","release_date":"2025-12-01","context_window":163840,"max_output_tokens":65536,"model_type":"chat","tagline":"We introduce DeepSeek-V3.2, a next-generation foundation model designed to unify high computational efficiency with state-of-the-art reasoning and agentic performance.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.269,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"deepseek/deepseek-v4-flash","name":"Deepseek V4 Flash","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Flash is a lightweight model meticulously designed by DeepSeek to deliver the ultimate combination of lightning-fast response times and unmatched cost-effectiveness.","status":"active","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":393216,"model_type":"chat","tagline":"DeepSeek-V4-Flash is a lightweight model meticulously designed by DeepSeek to deliver the ultimate combination of lightning-fast response times and unmatched cost-effectiveness.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.14,"output":0.28},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"deepseek/deepseek-v4-pro","name":"Deepseek V4 Pro","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Pro is the next-generation flagship open-source large language model developed by DeepSeek, delivering comprehensive performance that rivals the world's premier closed-source models.","status":"active","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":393216,"model_type":"chat","tagline":"DeepSeek-V4-Pro is the next-generation flagship open-source large language model developed by DeepSeek, delivering comprehensive performance that rivals the world's premier closed-source models.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.6,"output":3.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"mit","open_weight":true},{"id":"deepseek/deepseek_v3","name":"DeepSeek V3","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","status":"active","release_date":"2025-01-15","context_window":64000,"max_output_tokens":16000,"model_type":"chat","tagline":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.89,"output":0.89},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"dev/glm46","name":"dev/glm46","created_by":"dev","source":"official","family":"glm","status":"active","release_date":"2025-11-20","context_window":256000,"max_output_tokens":256000,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"elephant","name":"Ling-2.6-flash","created_by":"novita","source":"official","status":"active","release_date":"2026-04-10","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"google/gemma-3-12b-it","name":"Gemma3 12B","created_by":"google","source":"official","family":"gemma-3","status":"active","release_date":"2025-09-04","context_window":131072,"max_output_tokens":8192,"model_type":"chat","parameters":12,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.1},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemma-3-27b-it","name":"Gemma 3 27B","created_by":"google","source":"official","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","status":"active","release_date":"2025-03-25","context_window":98304,"max_output_tokens":16384,"model_type":"chat","parameters":27,"tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","capabilities":{"vision":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.119,"output":0.2},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemma-4-26b-a4b-it","name":"Gemma 4 26B A4B","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 26B A4B is built for developers who need scalable performance without sacrificing core capabilities.Crucially, it retains the massive 256K-token context window of the 31B model, making it highly competitive for long-context RAG and processing extensive, image-rich document datasets.","status":"active","release_date":"2026-04-02","context_window":262144,"max_output_tokens":131072,"model_type":"chat","parameters":26,"active_parameters":4,"tagline":"Gemma 4 26B A4B is built for developers who need scalable performance without sacrificing core capabilities.Crucially, it retains the massive 256K-token context window of the 31B model, making it high","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.13,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"google/gemma-4-31b-it","name":"Gemma 4 31B","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 31B is engineered to tackle the most demanding enterprise workloads and complex reasoning tasks.","status":"active","release_date":"2026-04-02","context_window":262144,"max_output_tokens":131072,"model_type":"chat","parameters":31,"tagline":"Gemma 4 31B is engineered to tackle the most demanding enterprise workloads and complex reasoning tasks.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.14,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"gryphe/mythomax-l2-13b","name":"Mythomax L2 13B","created_by":"gryphe","source":"official","description":"The idea behind this merge is that each layer is composed of several tensors, which are in turn responsible for specific functions.","status":"active","release_date":"2024-04-25","context_window":4096,"max_output_tokens":3200,"model_type":"chat","parameters":13,"tagline":"The idea behind this merge is that each layer is composed of several tensors, which are in turn responsible for specific functions.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.09},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"gt-4p","name":"gt-4p","created_by":"novita","source":"official","description":"chatgpt-4o","status":"active","release_date":"2025-03-31","max_output_tokens":131072,"model_type":"chat","tagline":"chatgpt-4o","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"inclusionai/ling-2.6-1t","name":"Ling-2.6-1T","created_by":"inclusionai","source":"official","status":"active","release_date":"2026-04-23","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"inclusionai/ling-2.6-flash","name":"Ling-2.6-flash","created_by":"inclusionai","source":"official","status":"active","release_date":"2026-04-24","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"inclusionai/ring-2.6-1t","name":"Ring-2.6-1T","created_by":"inclusionai","source":"official","status":"active","release_date":"2026-05-08","context_window":262144,"max_output_tokens":65536,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","pricing":{"input":0.3,"output":2.5}},{"id":"kwaipilot/kat-coder-pro","name":"Kat Coder Pro","created_by":"kwaipilot","source":"official","description":"KAT-Coder-Pro V2 by KwaiKAT is a non-reasoning model optimized for agentic coding.","status":"active","release_date":"2026-01-05","context_window":256000,"max_output_tokens":128000,"model_type":"chat","tagline":"KAT-Coder-Pro V2 by KwaiKAT is a non-reasoning model optimized for agentic coding.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"meta-llama/llama-3-70b-instruct","name":"Llama3 70B Instruct","created_by":"meta","source":"official","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","status":"preview","release_date":"2024-04-25","context_window":8192,"max_output_tokens":8000,"model_type":"chat","parameters":70,"tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.51,"output":0.74},"endpoints":["chat_completions"],"last_updated":"2026-06-29"},{"id":"meta-llama/llama-3-8b-instruct","name":"Llama 3 8B Instruct","created_by":"meta","source":"official","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","status":"preview","release_date":"2024-04-25","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":8,"tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.04,"output":0.04},"endpoints":["chat_completions"],"last_updated":"2026-06-29"},{"id":"meta-llama/llama-3.1-8b-instruct","name":"Llama 3.1 8B Instruct","created_by":"meta","source":"official","family":"llama-3.1","description":"Meta's latest class of models, Llama 3.1, launched with a variety of sizes and configurations.","status":"active","release_date":"2024-07-24","context_window":16384,"max_output_tokens":16384,"model_type":"chat","parameters":8,"tagline":"Meta's latest class of models, Llama 3.1, launched with a variety of sizes and configurations.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.02,"output":0.05},"endpoints":["chat_completions"],"last_updated":"2026-05-11","knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta-llama/llama-3.2-1b-instruct","name":"Llama 3.2 1B Instruct\t","created_by":"meta","source":"official","family":"llama-3.2","description":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).","status":"active","release_date":"2024-11-26","context_window":131000,"max_output_tokens":32000,"model_type":"chat","parameters":1,"tagline":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).","capabilities":{"structured_output":true,"streaming":true,"json_mode":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.02,"output":0.02},"endpoints":["chat_completions"],"last_updated":"2026-06-03","knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta-llama/llama-3.2-3b-instruct","name":"Llama 3.2 3B Instruct","created_by":"meta","source":"official","family":"llama-3.2","description":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out)","status":"active","release_date":"2024-11-26","context_window":32768,"max_output_tokens":32000,"model_type":"chat","parameters":3,"tagline":"The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out)","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.05},"endpoints":["chat_completions"],"last_updated":"2026-05-11","knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta-llama/llama-3.3-70b-instruct","name":"Llama 3.3 70B Instruct","created_by":"meta","source":"official","family":"llama-3.3","description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","status":"active","release_date":"2024-12-07","context_window":131072,"max_output_tokens":120000,"model_type":"chat","parameters":70,"tagline":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.135,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","knowledge_cutoff":"2023-12","license":"llama3.3","open_weight":true},{"id":"meta-llama/llama-4-maverick-17b-128e-instruct-fp8","name":"Llama 4 Maverick Instruct","created_by":"meta","source":"official","family":"llama-4","description":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward pass (400B total).","status":"active","release_date":"2025-04-06","context_window":1048576,"max_output_tokens":8192,"model_type":"chat","parameters":17,"tagline":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forw","capabilities":{"vision":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.27,"output":0.85},"endpoints":["chat_completions"],"last_updated":"2026-05-11","license":"llama4","open_weight":true},{"id":"meta-llama/llama-4-scout-17b-16e-instruct","name":"Llama 4 Scout Instruct","created_by":"meta","source":"official","family":"llama-4","description":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","status":"active","release_date":"2025-04-06","context_window":131072,"max_output_tokens":131072,"model_type":"chat","parameters":17,"tagline":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","capabilities":{"vision":true,"reasoning":true,"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.18,"output":0.59},"endpoints":["chat_completions"],"last_updated":"2026-05-11","license":"llama4","open_weight":true},{"id":"microsoft/wizardlm-2-8x22b","name":"Wizardlm 2 8x22B","created_by":"microsoft","source":"official","description":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","status":"active","release_date":"2024-04-24","context_window":65535,"max_output_tokens":8000,"model_type":"chat","tagline":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.62,"output":0.62},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"minimax/m2-her","name":"M2-her","created_by":"minimax","source":"official","description":"M2-her测试","status":"active","release_date":"2026-01-22","context_window":32000,"model_type":"chat","tagline":"M2-her测试","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2.1","name":"Minimax M2.1","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.1 is a cutting-edge AI model designed to revolutionize how developers build software.","status":"active","release_date":"2025-12-23","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax M2.1 is a cutting-edge AI model designed to revolutionize how developers build software.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5-highspeed","name":"MiniMax M2.5-highspeed","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.5-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","status":"active","release_date":"2026-03-02","context_window":204800,"max_output_tokens":131100,"model_type":"chat","tagline":"MiniMax M2.5-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5","name":"MiniMax M2.5","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","status":"active","release_date":"2026-02-12","context_window":204800,"max_output_tokens":131100,"model_type":"chat","tagline":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.7-highspeed","name":"MiniMax M2.7-highspeed","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.7-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","status":"active","release_date":"2026-04-02","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax M2.7-highspeed is an accelerated SOTA model engineered for scenarios demanding extreme efficiency.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":2.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2.7","name":"MiniMax M2.7","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M2.7 is an all-around evolved, versatile open-source large language model that seamlessly blends hardcore engineering productivity with high-EQ, human-like interaction capabilities.","status":"active","release_date":"2026-03-17","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax M2.7 is an all-around evolved, versatile open-source large language model that seamlessly blends hardcore engineering productivity with high-EQ, human-like interaction capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2","name":"MiniMax-M2","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","status":"active","release_date":"2025-10-27","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m3","name":"MiniMax M3","created_by":"minimax","source":"official","family":"minimax","status":"active","release_date":"2026-06-01","context_window":1000000,"max_output_tokens":131072,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","license":"proprietary","open_weight":false},{"id":"minimaxai/minimax-m1-80k","name":"MiniMax M1","created_by":"minimaxai","source":"official","family":"minimax","description":"MiniMax-M1: The World's First Open-Weight, Large-Scale Hybrid Attention Inference Model\n\nMiniMax-M1 adopts a Mixture of Experts (MoE) architecture and integrates the Flash Attention mechanism. The mod","status":"active","release_date":"2025-06-17","context_window":1000000,"max_output_tokens":40000,"model_type":"chat","tagline":"MiniMax-M1: The World's First Open-Weight, Large-Scale Hybrid Attention Inference Model\n\nMiniMax-M1 adopts a Mixture of Experts (MoE) architecture and integrates the Flash Attention mechanism. The mod","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"mistralai/mistral-nemo","name":"Mistral Nemo","created_by":"mistral","source":"official","family":"mistral-nemo","description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","status":"active","release_date":"2024-07-30","context_window":60288,"max_output_tokens":16000,"model_type":"chat","tagline":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.04,"output":0.17},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"apache-2.0","open_weight":true},{"id":"moonshotai/kimi-k2-0905","name":"Kimi K2 0905","created_by":"moonshot","source":"official","family":"kimi-k2","description":"Kimi K2 0905 is the September update of Kimi K2 0711.","status":"active","release_date":"2025-09-05","context_window":262144,"max_output_tokens":100352,"model_type":"chat","tagline":"Kimi K2 0905 is the September update of Kimi K2 0711.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"moonshotai/kimi-k2-instruct","name":"Kimi K2 Instruct","created_by":"moonshot","source":"official","family":"kimi-k2","description":"Kimi K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters.","status":"active","release_date":"2025-07-11","context_window":131072,"max_output_tokens":100352,"model_type":"chat","tagline":"Kimi K2 is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.57,"output":2.3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"moonshotai/kimi-k2-thinking","name":"Kimi K2 Thinking","created_by":"moonshot","source":"official","family":"kimi-k2","description":"The kimi-k2-thinking model is a general-purpose agentic reasoning model developed by Moonshot AI.","status":"active","release_date":"2025-11-07","context_window":262144,"max_output_tokens":100352,"model_type":"chat","tagline":"The kimi-k2-thinking model is a general-purpose agentic reasoning model developed by Moonshot AI.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-29","license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.5","name":"Kimi K2.5","created_by":"moonshot","source":"official","family":"kimi-k2.5","description":"Kimi K2.5 is the latest flagship iteration of Moonshot AI's large language model series, representing a significant leap in multimodal and agentic capabilities.","status":"active","release_date":"2026-01-27","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.5 is the latest flagship iteration of Moonshot AI's large language model series, representing a significant leap in multimodal and agentic capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.6","name":"Kimi K2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is an open-source, native multimodal agentic model that significantly advances practical capabilities in long-horizon coding, coding-driven design, and swarm-based task orchestration.","status":"active","release_date":"2026-04-20","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.6 is an open-source, native multimodal agentic model that significantly advances practical capabilities in long-horizon coding, coding-driven design, and swarm-based task orchestration.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.8,"output":3.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"moonshotai/kimi-k2.7-code","name":"Kimi K2.7 Code","created_by":"moonshot","source":"official","family":"kimi-k2.7","description":"Kimi K2.7 Code is MoonshotAI's strongest coding & agentic model — a 1T-parameter MoE (32B activated) , 256K context and interleaved thinking with multi-step tool calling.","status":"active","release_date":"2026-06-12","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.7 Code is MoonshotAI's strongest coding & agentic model — a 1T-parameter MoE (32B activated) , 256K context and interleaved thinking with multi-step tool calling.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.95,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"nex-agi/nex-n2-pro","name":"Nex-N2-Pro","created_by":"nex-agi","source":"official","description":"Nex-N2-Pro is Nex AGI's open-source agentic reasoning model, post-trained on Qwen3.5-397B-A17B (MoE).","status":"preview","release_date":"2026-06-04","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Nex-N2-Pro is Nex AGI's open-source agentic reasoning model, post-trained on Qwen3.5-397B-A17B (MoE).","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"nousresearch/hermes-2-pro-llama-3-8b","name":"Hermes 2 Pro Llama 3 8B","created_by":"nousresearch","source":"official","description":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mod","status":"active","release_date":"2024-06-27","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":8,"tagline":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mod","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.14,"output":0.14},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"nousresearch/nous-hermes-llama2-13b","name":"Nous Hermes Llama2 13B","created_by":"nousresearch","source":"official","description":"Nous-Hermes-Llama2-13b is a state-of-the-art language model fine-tuned on over 300,000 instructions.","status":"preview","release_date":"2024-04-25","context_window":4096,"max_output_tokens":32768,"model_type":"chat","parameters":13,"tagline":"Nous-Hermes-Llama2-13b is a state-of-the-art language model fine-tuned on over 300,000 instructions.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.17,"output":0.17},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"Nemotron 3 Nano 30B A3B","created_by":"nvidia","source":"official","description":"Nemotron-3-Nano-30B-A3B is NVIDIA's compute-efficient, open-weight reasoning model built for agentic AI.","status":"active","release_date":"2026-06-09","context_window":262144,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"Nemotron-3-Nano-30B-A3B is NVIDIA's compute-efficient, open-weight reasoning model built for agentic AI.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"openai/gpt-oss-120b","name":"OpenAI GPT OSS 120B","created_by":"openai","source":"official","family":"gpt-oss","description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","status":"active","release_date":"2025-08-06","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":120,"tagline":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":false,"vision":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.25},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b","name":"OpenAI: GPT OSS 20B","created_by":"openai","source":"official","family":"gpt-oss","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","status":"active","release_date":"2025-08-06","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":20,"tagline":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":false,"vision":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.04,"output":0.15},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openchat/openchat-7b","name":"OpenChat 7B","created_by":"openchat","source":"official","description":"OpenChat 7B is a library of open-source language models, fine-tuned with \"C-RLFT (Conditioned Reinforcement Learning Fine-Tuning)\" - a strategy inspired by offline reinforcement learning.","status":"preview","release_date":"2024-07-26","context_window":4096,"max_output_tokens":4096,"model_type":"chat","parameters":7,"tagline":"OpenChat 7B is a library of open-source language models, fine-tuned with \"C-RLFT (Conditioned Reinforcement Learning Fine-Tuning)\" - a strategy inspired by offline reinforcement learning.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.06,"output":0.06},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"paddlepaddle/paddleocr-vl","name":"PaddleOCR-VL","created_by":"paddlepaddle","source":"official","status":"active","release_date":"2025-10-22","context_window":16384,"max_output_tokens":16384,"model_type":"chat","capabilities":{"vision":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.02,"output":0.02},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen-2-7b-instruct","name":"Qwen 2 7B Instruct","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen2 is the newest series in the Qwen large language model family.","status":"preview","release_date":"2024-09-18","context_window":32768,"max_output_tokens":32768,"model_type":"chat","parameters":7,"tagline":"Qwen2 is the newest series in the Qwen large language model family.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.054,"output":0.054},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"qwen/qwen-2-vl-72b-instruct","name":"Qwen 2 VL 72B Instruct","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen2 VL 72B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\nSoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on ","status":"preview","release_date":"2024-12-19","context_window":32768,"max_output_tokens":120000,"model_type":"chat","parameters":72,"tagline":"Qwen2 VL 72B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\nSoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on ","capabilities":{"vision":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.45,"output":0.45},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"qwen/qwen-2.5-72b-instruct","name":"Qwen 2.5 72B Instruct","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen2.5 is the latest series of Qwen large language models.","status":"active","release_date":"2024-10-15","context_window":32000,"max_output_tokens":8192,"model_type":"chat","parameters":72,"tagline":"Qwen2.5 is the latest series of Qwen large language models.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.38,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen-mt-plus","name":"Qwen MT Plus","created_by":"alibaba","source":"official","family":"qwen","description":"Qwen-MT is a large language model optimized for machine translation, built upon the foundation of the Tongyi Qianwen model.","status":"active","release_date":"2025-09-03","context_window":16384,"max_output_tokens":8192,"model_type":"chat","tagline":"Qwen-MT is a large language model optimized for machine translation, built upon the foundation of the Tongyi Qianwen model.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.25,"output":0.75},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen2.5-7b-instruct","name":"Qwen2.5 7B Instruct","created_by":"alibaba","source":"official","family":"qwen2.5","description":"Qwen2.5 is the latest series of Qwen large language models.","status":"preview","release_date":"2025-04-16","context_window":32000,"max_output_tokens":8192,"model_type":"chat","parameters":7,"tagline":"Qwen2.5 is the latest series of Qwen large language models.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.07},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen2.5-vl-72b-instruct","name":"Qwen2.5 VL 72B Instruct","created_by":"alibaba","source":"official","family":"qwen2.5","description":"Qwen2.5-VL, the latest vision-language model in the Qwen2.5 series, delivers enhanced multimodal capabilities including advanced visual comprehension for object/text recognition, chart/layout analysis, and agent-based dynamic tool orchestration.","status":"preview","release_date":"2025-03-25","context_window":32768,"max_output_tokens":32768,"model_type":"chat","parameters":72,"tagline":"Qwen2.5-VL, the latest vision-language model in the Qwen2.5 series, delivers enhanced multimodal capabilities including advanced visual comprehension for object/text recognition, chart/layout analysis","capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.8,"output":0.8},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3-235b-a22b-fp8","name":"Qwen3 235B A22B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of inference and non-inference modes, enabling seamless switching between modes during conversations.","status":"active","release_date":"2025-04-29","context_window":40960,"max_output_tokens":20000,"model_type":"chat","parameters":235,"active_parameters":22,"tagline":"Achieves effective integration of inference and non-inference modes, enabling seamless switching between modes during conversations.","capabilities":{"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.2,"output":0.8},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-235b-a22b-instruct-2507","name":"Qwen3 235B A22B Instruct 2507","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","status":"active","release_date":"2025-07-22","context_window":131072,"max_output_tokens":16384,"model_type":"chat","parameters":235,"active_parameters":22,"tagline":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.58},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-235b-a22b-thinking-2507","name":"Qwen3 235B A22b Thinking 2507","created_by":"alibaba","source":"official","family":"qwen3","description":"The Qwen3-235B-A22B-Thinking-2507 represents the newest thinking-enabled model in the Qwen3 series, delivering groundbreaking improvements in reasoning capabilities.","status":"active","release_date":"2025-07-25","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":235,"active_parameters":22,"tagline":"The Qwen3-235B-A22B-Thinking-2507 represents the newest thinking-enabled model in the Qwen3 series, delivering groundbreaking improvements in reasoning capabilities.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":3},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-30b-a3b-fp8","name":"Qwen3 30B A3B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","status":"active","release_date":"2025-04-29","context_window":40960,"max_output_tokens":20000,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.45},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-32b-fp8","name":"Qwen3 32B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","status":"active","release_date":"2025-04-29","context_window":40960,"max_output_tokens":20000,"model_type":"chat","parameters":32,"tagline":"Achieves effective integration of inference and non-inference modes, allowing seamless switching between modes during conversations.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.45},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-4b-fp8","name":"Qwen3 4B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless switching during conversations.","status":"preview","release_date":"2025-04-29","context_window":128000,"max_output_tokens":8192,"model_type":"chat","parameters":4,"tagline":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless switching during conversations.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3-8b-fp8","name":"Qwen3 8B","created_by":"alibaba","source":"official","family":"qwen3","description":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless mode switching during conversations.","status":"active","release_date":"2025-04-29","context_window":128000,"max_output_tokens":20000,"model_type":"chat","parameters":8,"tagline":"Achieves effective integration of reasoning and non-reasoning modes, allowing seamless mode switching during conversations.","capabilities":{"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.035,"output":0.138},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-coder-30b-a3b-instruct","name":"Qwen3 Coder 30b A3B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.","status":"active","release_date":"2025-10-09","context_window":160000,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, an","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.27},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-coder-480b-a35b-instruct","name":"Qwen3 Coder 480B A35B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Coder-480B-A35B-Instruct is a cutting-edge open coding model from Qwen, matching Claude Sonnet’s performance in agentic programming, browser automation, and core development tasks.","status":"active","release_date":"2025-07-23","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":480,"active_parameters":35,"tagline":"Qwen3-Coder-480B-A35B-Instruct is a cutting-edge open coding model from Qwen, matching Claude Sonnet’s performance in agentic programming, browser automation, and core development tasks.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.38,"output":1.55},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3-coder-next","name":"Qwen3 Coder Next","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Coder-Next is an open-weight language model specifically engineered for coding agents and local development environments.","status":"active","release_date":"2026-02-03","context_window":262144,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3-Coder-Next is an open-weight language model specifically engineered for coding agents and local development environments.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.2,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-max","name":"Qwen3 Max","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen/qwen3-max, Enhanced with specialized upgrades in agent programming and tool calling.","status":"active","release_date":"2025-09-24","context_window":262144,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen/qwen3-max, Enhanced with specialized upgrades in agent programming and tool calling.","capabilities":{"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.11,"output":8.45},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"Qwen3 Next 80B A3B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","status":"active","release_date":"2025-09-10","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":80,"active_parameters":3,"tagline":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"Qwen3 Next 80B A3B Thinking","created_by":"alibaba","source":"official","family":"qwen3","description":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","status":"active","release_date":"2025-09-10","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":80,"active_parameters":3,"tagline":"Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-omni-30b-a3b-instruct","name":"Qwen3 Omni 30B A3B Instruct","created_by":"alibaba","source":"official","family":"qwen3","description":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","status":"active","release_date":"2025-09-24","context_window":65536,"max_output_tokens":16384,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","pricing":{"input":0.25,"output":0.97}},{"id":"qwen/qwen3-omni-30b-a3b-thinking","name":"Qwen3 Omni 30B A3B Thinking","created_by":"alibaba","source":"official","family":"qwen3","description":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","status":"active","release_date":"2025-09-24","context_window":65536,"max_output_tokens":16384,"model_type":"chat","parameters":30,"active_parameters":3,"tagline":"The Qwen-Omni model accepts combined inputs of text and a single additional modality (image, audio, or video) to generate responses in text or speech.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","pricing":{"input":0.25,"output":0.97}},{"id":"qwen/qwen3-vl-235b-a22b-instruct","name":"Qwen3 VL 235B A22B Instruct","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-09-24","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":235,"active_parameters":22,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":1.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-235b-a22b-thinking","name":"Qwen3 VL 235B A22B Thinking","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-09-24","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":235,"active_parameters":22,"capabilities":{"vision":true,"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.98,"output":3.95},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-30b-a3b-instruct","name":"qwen/qwen3-vl-30b-a3b-instruct","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-10-11","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.7},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-30b-a3b-thinking","name":"qwen/qwen3-vl-30b-a3b-thinking","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-10-11","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":30,"active_parameters":3,"capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":1},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-vl-8b-instruct","name":"qwen/qwen3-vl-8b-instruct","created_by":"alibaba","source":"official","family":"qwen3","status":"active","release_date":"2025-10-17","context_window":131072,"max_output_tokens":32768,"model_type":"chat","parameters":8,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.08,"output":0.5},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-122b-a10b","name":"Qwen3.5-122B-A10B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5-122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","status":"active","release_date":"2026-02-26","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":122,"active_parameters":10,"tagline":"The Qwen3.5-122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference eff","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.4,"output":3.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-27b","name":"Qwen3.5-27B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","status":"active","release_date":"2026-02-26","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":27,"tagline":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-35b-a3b","name":"Qwen3.5-35B-A3B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.","status":"active","release_date":"2026-02-26","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":35,"active_parameters":3,"tagline":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inf","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen3.5-397B-A17B","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 series 397B-A17B native vision-language model is based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher inference efficiency.","status":"active","release_date":"2026-02-16","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":397,"active_parameters":17,"tagline":"The Qwen3.5 series 397B-A17B native vision-language model is based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher in","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":3.6},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3.5-plus","name":"Qwen3.5-Plus","created_by":"alibaba","source":"official","family":"qwen3.5","description":"The Qwen3.5 native vision-language series Plus models are based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher inference efficiency.","status":"active","release_date":"2026-02-16","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"The Qwen3.5 native vision-language series Plus models are based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher infer","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.6-27b","name":"Qwen3.6-27B","created_by":"alibaba","source":"official","family":"qwen3.6","description":"The Qwen3.6 27B native vision-language dense model builds upon the 3.5-27B version, with key improvements in agentic coding capabilities and enhanced STEM reasoning and inference skills.","status":"active","release_date":"2026-04-23","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":27,"tagline":"The Qwen3.6 27B native vision-language dense model builds upon the 3.5-27B version, with key improvements in agentic coding capabilities and enhanced STEM reasoning and inference skills.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":3.6},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen3.6-35B-A3B","created_by":"alibaba","source":"official","family":"qwen3.6","description":"The Qwen3.6 35B-A3B native vision-language model is built on a hybrid architecture that integrates linear attention mechanisms with a sparse mixture-of-experts framework, achieving higher inference efficiency.","status":"active","release_date":"2026-04-21","context_window":262144,"max_output_tokens":65536,"model_type":"chat","parameters":35,"active_parameters":3,"tagline":"The Qwen3.6 35B-A3B native vision-language model is built on a hybrid architecture that integrates linear attention mechanisms with a sparse mixture-of-experts framework, achieving higher inference ef","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.248,"output":1.485},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.6-plus","name":"Qwen3.6-Plus","created_by":"alibaba","source":"official","family":"qwen3.6","description":"The Qwen3.6 native vision-language Plus series models demonstrate exceptional performance on par with the current state-of-the-art models, with a significant improvement in overall results compared to the 3.5 series.","status":"active","release_date":"2026-04-21","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"The Qwen3.6 native vision-language Plus series models demonstrate exceptional performance on par with the current state-of-the-art models, with a significant improvement in overall results compared to","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen/qwen3.7-max","name":"Qwen3.7-Max","created_by":"alibaba","source":"official","family":"qwen3.7","description":"The Qwen3.7 Max model—the largest and most powerful in the series—is currently available for users to try out its text-only features.","status":"active","release_date":"2026-05-21","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"The Qwen3.7 Max model—the largest and most powerful in the series—is currently available for users to try out its text-only features.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.25,"output":3.75},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"sao10k/l3-70b-euryale-v2.1","name":"L3 70B Euryale V2.1\t","created_by":"sao10k","source":"official","description":"The uncensored llama3 model is a powerhouse of creativity, excelling in both roleplay and story writing.","status":"active","release_date":"2024-06-18","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":70,"tagline":"The uncensored llama3 model is a powerhouse of creativity, excelling in both roleplay and story writing.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.48,"output":1.48},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"sao10k/l3-8b-lunaris","name":"Sao10k L3 8B Lunaris\t","created_by":"sao10k","source":"official","description":"A generalist / roleplaying model merge based on Llama 3.","status":"active","release_date":"2024-11-28","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":8,"tagline":"A generalist / roleplaying model merge based on Llama 3.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.05},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"Sao10K/L3-8B-Stheno-v3.2","name":"L3 8B Stheno V3.2","created_by":"sao10k","source":"official","description":"Sao10K/L3-8B-Stheno-v3.2 is a highly skilled actor that excels at fully immersing itself in any role assigned.","status":"active","release_date":"2024-11-29","context_window":8192,"max_output_tokens":32000,"model_type":"chat","parameters":8,"tagline":"Sao10K/L3-8B-Stheno-v3.2 is a highly skilled actor that excels at fully immersing itself in any role assigned.","capabilities":{"tool_call":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.05},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"sao10k/l31-70b-euryale-v2.2","name":"L31 70B Euryale V2.2","created_by":"sao10k","source":"official","description":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from Sao10k.","status":"active","release_date":"2024-09-19","context_window":8192,"max_output_tokens":8192,"model_type":"chat","parameters":70,"tagline":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from Sao10k.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.48,"output":1.48},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"stepfun/step-3.7-flash","name":"Step 3.7 Flash","created_by":"stepfun","source":"official","description":"Step-3.7-Flash is StepFun's flagship high-efficiency multimodal reasoning model, built on a sparse Mixture-of-Experts architecture (198B total / ~11B active parameters) that delivers high throughput and low latency for real-time agentic workflows and high-frequency calls.","status":"active","release_date":"2026-06-12","context_window":262144,"max_output_tokens":256000,"model_type":"chat","tagline":"Step-3.7-Flash is StepFun's flagship high-efficiency multimodal reasoning model, built on a sparse Mixture-of-Experts architecture (198B total / ~11B active parameters) that delivers high throughput a","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":1.15},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"teknium/openhermes-2.5-mistral-7b","name":"Openhermes2.5 Mistral 7B","created_by":"teknium","source":"official","description":"OpenHermes 2.5 Mistral 7B is a state of the art Mistral Fine-tune, a continuation of OpenHermes 2 model, which trained on additional code datasets.","status":"preview","release_date":"2024-04-24","context_window":4096,"max_output_tokens":8000,"model_type":"chat","parameters":7,"tagline":"OpenHermes 2.5 Mistral 7B is a state of the art Mistral Fine-tune, a continuation of OpenHermes 2 model, which trained on additional code datasets.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.17,"output":0.17},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"thudm/glm-4-32b-0414","name":"GLM-4-32B-0414","created_by":"thudm","source":"official","family":"glm-4","description":"GLM-4-32B-0414 is the latest open-source model in the GLM series, featuring 32 billion parameters.","status":"active","release_date":"2025-04-16","context_window":32000,"max_output_tokens":32000,"model_type":"chat","parameters":32,"tagline":"GLM-4-32B-0414 is the latest open-source model in the GLM series, featuring 32 billion parameters.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":1.66},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"xiaomimimo/mimo-v2-flash","name":"XiaomiMiMo/MiMo-V2-Flash","created_by":"xiaomi","source":"official","family":"mimo","description":"Xiaomi MiMo-V2-Flash is a proprietary MoE model developed by Xiaomi, designed for extreme inference efficiency with 309B total parameters (15B active).","status":"preview","release_date":"2025-12-19","context_window":262144,"max_output_tokens":32000,"model_type":"chat","tagline":"Xiaomi MiMo-V2-Flash is a proprietary MoE model developed by Xiaomi, designed for extreme inference efficiency with 309B total parameters (15B active).","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.11,"output":0.33},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10","knowledge_cutoff":"2024-12","license":"mit","parameters":309,"active_parameters":15,"architecture":"moe","open_weight":true},{"id":"xiaomimimo/mimo-v2-pro","name":"XiaomiMiMo/MiMo-V2-Pro","created_by":"xiaomi","source":"official","family":"mimo","description":"Xiaomi’s MiMo-V2-Pro is a flagship, >1T parameter foundation model with a 1M context window, specifically optimized for advanced agentic workflows.","status":"preview","release_date":"2026-03-20","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"Xiaomi’s MiMo-V2-Pro is a flagship, >1T parameter foundation model with a 1M context window, specifically optimized for advanced agentic workflows.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2,"output":6},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"proprietary","parameters":1000,"active_parameters":42,"architecture":"moe","open_weight":false},{"id":"xiaomimimo/mimo-v2.5-pro","name":"XiaomiMiMo/MiMo-V2.5-Pro","created_by":"xiaomi","source":"official","family":"mimo","description":"MiMo-V2.5-Pro is purpose-built to push the boundaries of complex software engineering and extreme long-horizon tasks.","status":"active","release_date":"2026-05-06","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"MiMo-V2.5-Pro is purpose-built to push the boundaries of complex software engineering and extreme long-horizon tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.522,"output":1.044},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"xiaomimimo/mimo-v2.5","name":"XiaomiMiMo/MiMo-V2.5","created_by":"xiaomi","source":"official","family":"mimo","description":"Xiaomi's native omni-modal model (text / image / video / audio understanding) with Pro-level agentic performance at ~half the inference cost and a 1M-token context — built for cost-efficient, percepti","status":"active","release_date":"2026-05-29","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"mimo-v2.5, a native omni-modal model supporting text, image, video, and audio understanding, equipped with powerful agentic capabilities.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.168,"output":0.336},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"zai-org/autoglm-phone-9b-multilingual","name":"AutoGLM-Phone-9B-Multilingual","created_by":"zai","source":"official","description":"Phone Agent is a mobile intelligent assistant framework built on AutoGLM, capable of understanding smartphone screens through multimodal perception and executing automated operations to complete tasks.","status":"active","release_date":"2025-12-10","context_window":65536,"max_output_tokens":65536,"model_type":"chat","parameters":9,"tagline":"Phone Agent is a mobile intelligent assistant framework built on AutoGLM, capable of understanding smartphone screens through multimodal perception and executing automated operations to complete tasks","capabilities":{"vision":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.035,"output":0.138},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"zai-org/glm-4.5-air","name":"zai-org/glm-4.5-air","created_by":"zai","source":"official","family":"glm-4.5","status":"active","release_date":"2025-10-13","context_window":131072,"max_output_tokens":98304,"model_type":"chat","capabilities":{"tool_call":true,"reasoning":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.13,"output":0.85},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11"},{"id":"zai-org/glm-4.5","name":"GLM-4.5","created_by":"zai","source":"official","family":"glm-4.5","description":"GLM-4.5 Series Models are foundation models specifically engineered for intelligent agents.","status":"preview","release_date":"2025-07-28","context_window":131072,"max_output_tokens":98304,"model_type":"chat","tagline":"GLM-4.5 Series Models are foundation models specifically engineered for intelligent agents.","capabilities":{"tool_call":true,"reasoning":true,"streaming":true,"structured_output":true,"prompt_caching":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-29","license":"mit","parameters":355,"active_parameters":32,"architecture":"moe","open_weight":true},{"id":"zai-org/glm-4.5v","name":"GLM 4.5V","created_by":"zai","source":"official","family":"glm-4.5","description":"Z.ai's GLM-4.5V sets a new standard in visual reasoning, achieving SOTA performance across 42 benchmarks among open-source models.","status":"active","release_date":"2025-08-11","context_window":65536,"max_output_tokens":16384,"model_type":"chat","tagline":"Z.ai's GLM-4.5V sets a new standard in visual reasoning, achieving SOTA performance across 42 benchmarks among open-source models.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.6,"output":1.8},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","parameters":106,"active_parameters":12,"architecture":"moe","open_weight":true},{"id":"zai-org/glm-4.6","name":"GLM 4.6","created_by":"zai","source":"official","family":"glm-4.6","description":"As the latest iteration in the GLM series, GLM-4.6 achieves comprehensive enhancements across multiple domains, including real-world coding, long-context processing, reasoning, searching, writing, and agentic applications.","status":"active","release_date":"2025-09-30","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"As the latest iteration in the GLM series, GLM-4.6 achieves comprehensive enhancements across multiple domains, including real-world coding, long-context processing, reasoning, searching, writing, and","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-4.6v","name":"GLM 4.6V","created_by":"zai","source":"official","family":"glm-4.6","description":"GLM-4.6V represents a significant multimodal advancement in the GLM series, featuring a 128k-token training context window and achieving state-of-the-art visual understanding accuracy for models of its parameter scale.","status":"active","release_date":"2025-12-08","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"GLM-4.6V represents a significant multimodal advancement in the GLM series, featuring a 128k-token training context window and achieving state-of-the-art visual understanding accuracy for models of it","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":0.9},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-4.7-flash","name":"GLM-4.7-Flash","created_by":"zai","source":"official","family":"glm-4.7","description":"GLM-4.7-Flash, a state-of-the-art model in the 30B class, delivers a compelling balance of high performance and efficiency.","status":"active","release_date":"2026-01-19","context_window":200000,"max_output_tokens":128000,"model_type":"chat","tagline":"GLM-4.7-Flash, a state-of-the-art model in the 30B class, delivers a compelling balance of high performance and efficiency.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.07,"output":0.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-4.7-h","name":"GLM-4.7","created_by":"zai","source":"official","family":"glm-4.7","description":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","status":"active","release_date":"2026-01-26","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"zai-org/glm-4.7","name":"GLM-4.7","created_by":"zai","source":"official","family":"glm-4.7","description":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","status":"active","release_date":"2025-12-22","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-4.7 is Z.AI's latest flagship model, with major upgrades focused on advanced coding capabilities and more reliable multi-step reasoning and execution.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","open_weight":true},{"id":"zai-org/glm-5-turbo","name":"GLM-5-Turbo","created_by":"zai","source":"official","family":"glm-5","description":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","status":"active","release_date":"2026-03-26","context_window":202800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.2,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false},{"id":"zai-org/glm-5.1","name":"GLM-5.1","created_by":"zai","source":"official","family":"glm-5.1","description":"GLM-5.1 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","status":"active","release_date":"2026-04-07","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5.1 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.38,"output":4.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"mit","open_weight":true},{"id":"zai-org/glm-5.2","name":"GLM 5.2","created_by":"zai","source":"official","family":"glm-5.2","description":"GLM-5.2 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","status":"active","release_date":"2026-06-12","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5.2 is Z.AI's latest flagship model, meticulously engineered for long-horizon autonomous tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.4,"output":4.4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-18","license":"mit","open_weight":true},{"id":"zai-org/glm-5","name":"GLM-5","created_by":"zai","source":"official","family":"glm-5","description":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","status":"active","release_date":"2026-02-08","context_window":202800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5 is an open-source foundation model engineered for complex system engineering and long-horizon Agent tasks, delivering reliable productivity for top-tier programmers.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1,"output":3.2},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-05-11","license":"mit","parameters":355,"active_parameters":32,"open_weight":true},{"id":"zai-org/glm-5v-turbo","name":"GLM-5V-Turbo","created_by":"zai","source":"official","family":"glm-5","description":"GLM-5V-Turbo is Z.AI’s first multimodal coding foundation model, built for vision-based coding tasks.","status":"active","release_date":"2026-04-02","context_window":204800,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5V-Turbo is Z.AI’s first multimodal coding foundation model, built for vision-based coding tasks.","capabilities":{"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"streaming":true,"json_mode":true,"prompt_caching":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.2,"output":4},"endpoints":["chat_completions"],"tools":["function_calling"],"last_updated":"2026-06-03","license":"mit","open_weight":true},{"id":"zai-org/glm-ocr","name":"GLM-OCR","created_by":"zai","source":"official","family":"glm","description":"GLM-OCR is a lightweight professional OCR model with only 0.9B parameters, achieving SOTA performance with a score of 94.62 on OmniDocBench V1.5.","status":"active","release_date":"2026-02-04","context_window":32000,"max_output_tokens":32000,"model_type":"chat","tagline":"GLM-OCR is a lightweight professional OCR model with only 0.9B parameters, achieving SOTA performance with a score of 94.62 on OmniDocBench V1.5.","capabilities":{"structured_output":true,"streaming":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.03},"endpoints":["chat_completions"],"last_updated":"2026-06-03","license":"proprietary","parameters":0.9,"open_weight":false}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/nvidia.ts b/packages/npm/src/providers/nvidia.ts
index 960f26978..e321d8d41 100644
--- a/packages/npm/src/providers/nvidia.ts
+++ b/packages/npm/src/providers/nvidia.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"nvidia","name":"NVIDIA","region":"US","headquarters":"Santa Clara, CA","founded":1993,"url":"https://build.nvidia.com","api_url":"https://integrate.api.nvidia.com/v1","docs_url":"https://docs.api.nvidia.com","pricing_url":"https://build.nvidia.com/pricing","description":"NVIDIA NIM provides optimized inference for 190+ models on NVIDIA GPU infrastructure.","type":"cloud","playground_url":"https://build.nvidia.com","sdk":{"python":"openai"},"openai_compatible":true,"free_tier":true,"models_url":"https://build.nvidia.com/explore/discover","twitter_url":"https://x.com/nvidia","blog_url":"https://blogs.nvidia.com","terms_url":"https://developer.nvidia.com/legal/terms","support_url":"https://forums.developer.nvidia.com","icon":"","models":[{"id":"01-ai/yi-large","name":"yi-large","created_by":"01-ai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"abacusai/dracarys-llama-3.1-70b-instruct","name":"dracarys-llama-3.1-70b-instruct","created_by":"abacusai","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":70,"tools":["function_calling"]},{"id":"adept/fuyu-8b","name":"fuyu-8b","created_by":"adept","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":8},{"id":"ai21labs/jamba-1.5-large-instruct","name":"jamba-1.5-large-instruct","created_by":"ai21","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"ai21labs/jamba-1.5-mini-instruct","name":"jamba-1.5-mini-instruct","created_by":"ai21","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"aisingapore/sea-lion-7b-instruct","name":"sea-lion-7b-instruct","created_by":"aisingapore","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"baai/bge-m3","name":"bge-m3","created_by":"baai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"baichuan-inc/baichuan2-13b-chat","name":"baichuan2-13b-chat","created_by":"baichuan","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":13,"tools":["function_calling"]},{"id":"bigcode/starcoder2-15b","name":"starcoder2-15b","created_by":"bigcode","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":15},{"id":"bigcode/starcoder2-7b","name":"starcoder2-7b","created_by":"bigcode","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"bytedance/seed-oss-36b-instruct","name":"seed-oss-36b-instruct","created_by":"bytedance","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":36,"tools":["function_calling"]},{"id":"databricks/dbrx-instruct","name":"dbrx-instruct","created_by":"databricks","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"]},{"id":"deepseek-ai/deepseek-coder-6.7b-instruct","name":"deepseek-coder-6.7b-instruct","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true,"tool_call":true},"parameters":6.7,"tools":["function_calling"]},{"id":"deepseek-ai/deepseek-r1-distill-llama-8b","name":"deepseek-r1-distill-llama-8b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-r1-distill-qwen-14b","name":"deepseek-r1-distill-qwen-14b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":14,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-r1-distill-qwen-32b","name":"deepseek-r1-distill-qwen-32b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":32,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-r1-distill-qwen-7b","name":"deepseek-r1-distill-qwen-7b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v3.1-terminus","name":"deepseek-v3.1-terminus","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v3.1","name":"deepseek-v3.1","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v3.2","name":"deepseek-v3.2","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v4-flash","name":"deepseek-v4-flash","created_by":"deepseek","source":"official","family":"deepseek","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","description":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","context_window":1000000,"max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"deepseek-ai/deepseek-v4-pro","name":"deepseek-v4-pro","created_by":"deepseek","source":"official","family":"deepseek","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","description":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","context_window":1000000,"max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"google/codegemma-1.1-7b","name":"codegemma-1.1-7b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"google/codegemma-7b","name":"codegemma-7b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"google/deplot","name":"deplot","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"google/diffusiongemma-26b-a4b-it","name":"diffusiongemma-26b-a4b-it","created_by":"google","source":"official","parameters":26,"active_parameters":4,"capabilities":{"streaming":true},"last_updated":"2026-06-18"},{"id":"google/gemma-2-27b-it","name":"gemma-2-27b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true},"model_type":"chat","parameters":27,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2-2b-it","name":"gemma-2-2b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true},"model_type":"chat","parameters":2,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2-9b-it","name":"gemma-2-9b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true},"model_type":"chat","parameters":9,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2b","name":"gemma-2b","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2b","capabilities":{"streaming":true},"model_type":"chat","parameters":2,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-12b-it","name":"gemma-3-12b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":12,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-1b-it","name":"gemma-3-1b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":1,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-27b-it","name":"gemma-3-27b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":27,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-4b-it","name":"gemma-3-4b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":4,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3n-e2b-it","name":"gemma-3n-e2b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3n","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3n-e4b-it","name":"gemma-3n-e4b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3n","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-4-31b-it","name":"gemma-4-31b-it","created_by":"google","source":"official","family":"gemma-4","parameters":31,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-03"},{"id":"google/gemma-7b","name":"gemma-7b","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-7b","capabilities":{"streaming":true},"model_type":"chat","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/paligemma","name":"paligemma","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"google/recurrentgemma-2b","name":"recurrentgemma-2b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":2},{"id":"google/shieldgemma-9b","name":"shieldgemma-9b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":9},{"id":"gotocompany/gemma-2-9b-cpt-sahabatai-instruct","name":"gemma-2-9b-cpt-sahabatai-instruct","created_by":"gotocompany","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":9,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-3.0-3b-a800m-instruct","name":"granite-3.0-3b-a800m-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":3,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-3.0-8b-instruct","name":"granite-3.0-8b-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-3.3-8b-instruct","name":"granite-3.3-8b-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-34b-code-instruct","name":"granite-34b-code-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":34,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-8b-code-instruct","name":"granite-8b-code-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-guardian-3.0-8b","name":"granite-guardian-3.0-8b","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"moderation","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"igenius/colosseum_355b_instruct_16k","name":"colosseum_355b_instruct_16k","created_by":"igenius","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":355,"tools":["function_calling"]},{"id":"igenius/italia_10b_instruct_16k","name":"italia_10b_instruct_16k","created_by":"igenius","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":10,"tools":["function_calling"]},{"id":"institute-of-science-tokyo/llama-3.1-swallow-70b-instruct-v0.1","name":"llama-3.1-swallow-70b-instruct-v0.1","created_by":"institute-of-science-tokyo","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"institute-of-science-tokyo/llama-3.1-swallow-8b-instruct-v0.1","name":"llama-3.1-swallow-8b-instruct-v0.1","created_by":"institute-of-science-tokyo","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"marin/marin-8b-instruct","name":"marin-8b-instruct","created_by":"marin","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"]},{"id":"mediatek/breeze-7b-instruct","name":"breeze-7b-instruct","created_by":"mediatek","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"meta/codellama-70b","name":"codellama-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"family":"codellama","model_type":"code","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.1-405b-instruct","name":"llama-3.1-405b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":405,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta/llama-3.1-70b-instruct","name":"llama-3.1-70b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta/llama-3.1-8b-instruct","name":"llama-3.1-8b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta/llama-3.2-11b-vision-instruct","name":"llama-3.2-11b-vision-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":11,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.2-1b-instruct","name":"llama-3.2-1b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.2-3b-instruct","name":"llama-3.2-3b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.2-90b-vision-instruct","name":"llama-3.2-90b-vision-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":90,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.3-70b-instruct","name":"llama-3.3-70b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.3","open_weight":true},{"id":"meta/llama-4-maverick-17b-128e-instruct","name":"llama-4-maverick-17b-128e-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":17,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":16384,"license":"llama4","open_weight":true},{"id":"meta/llama-4-scout-17b-16e-instruct","name":"llama-4-scout-17b-16e-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":17,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":16384,"license":"llama4","open_weight":true},{"id":"meta/llama-guard-4-12b","name":"llama-guard-4-12b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-guard","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"moderation","parameters":12,"modalities":{"input":["text"],"output":["text"]},"license":"llama4","open_weight":true},{"id":"meta/llama2-70b","name":"llama2-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama3-70b-instruct","name":"llama3-70b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama3-8b-instruct","name":"llama3-8b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/kosmos-2","name":"kosmos-2","created_by":"microsoft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"microsoft/phi-3-medium-128k-instruct","name":"phi-3-medium-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-medium-4k-instruct","name":"phi-3-medium-4k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-mini-128k-instruct","name":"phi-3-mini-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-mini-4k-instruct","name":"phi-3-mini-4k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-small-128k-instruct","name":"phi-3-small-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-small-8k-instruct","name":"phi-3-small-8k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-vision-128k-instruct","name":"phi-3-vision-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3.5-mini-instruct","name":"phi-3.5-mini-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3.5-moe-instruct","name":"phi-3.5-moe-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3.5-vision-instruct","name":"phi-3.5-vision-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-4-mini-flash-reasoning","name":"phi-4-mini-flash-reasoning","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-4-mini-instruct","name":"phi-4-mini-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-4-multimodal-instruct","name":"phi-4-multimodal-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"minimaxai/minimax-m2.1","name":"minimax-m2.1","created_by":"minimax","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","tagline":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","context_window":204800,"license":"mit","open_weight":true},{"id":"minimaxai/minimax-m2.5","name":"minimax-m2.5","created_by":"minimax","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Peak Performance. Ultimate Value. Master the Complex","tagline":"Peak Performance.","context_window":204800,"license":"mit","open_weight":true},{"id":"minimaxai/minimax-m2.7","name":"minimax-m2.7","created_by":"minimax","source":"official","family":"minimax","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-12","description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","context_window":204800,"license":"proprietary","open_weight":false},{"id":"minimaxai/minimax-m3","name":"minimax-m3","created_by":"minimax","source":"official","family":"minimax","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","context_window":1000000,"license":"proprietary","open_weight":false},{"id":"mistralai/codestral-22b-instruct-v0.1","name":"codestral-22b-instruct-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"codestral","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":22,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/devstral-2-123b-instruct-2512","name":"devstral-2-123b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","alias":"mistralai/devstral-2-123b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":123,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/devstral-2-123b-instruct","name":"devstral-2-123b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/devstral-2-123b-instruct-2512"],"model_type":"chat","parameters":123,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/magistral-small-2506","name":"magistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-03-23","alias":"mistralai/magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"description":"Our small reasoning model released July 2025.","tagline":"Our small reasoning model released July 2025.","context_window":40000,"license":"apache-2.0","open_weight":true},{"id":"mistralai/magistral-small","name":"magistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"snapshots":["mistralai/magistral-small-2506"],"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"description":"Our small reasoning model released July 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","context_window":40000,"license":"apache-2.0","open_weight":true},{"id":"mistralai/mamba-codestral-7b-v0.1","name":"mamba-codestral-7b-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mathstral-7b-v0.1","name":"mathstral-7b-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/ministral-14b-instruct-2512","name":"ministral-14b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","alias":"mistralai/ministral-14b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":14,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/ministral-14b-instruct","name":"ministral-14b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/ministral-14b-instruct-2512"],"model_type":"chat","parameters":14,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-7b-instruct-v0.2","name":"mistral-7b-instruct-v0.2","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-7b","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-7b-instruct-v0.3","name":"mistral-7b-instruct-v0.3","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-7b","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large-2-instruct","name":"mistral-large-2-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large-3-675b-instruct-2512","name":"mistral-large-3-675b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","alias":"mistralai/mistral-large-3-675b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":675,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large-3-675b-instruct","name":"mistral-large-3-675b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/mistral-large-3-675b-instruct-2512"],"model_type":"chat","parameters":675,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large","name":"mistral-large","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024.","context_window":32000,"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-medium-3-instruct","name":"mistral-medium-3-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-medium","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-medium-3.5-128b","name":"mistral-medium-3.5-128b","created_by":"mistral","source":"official","family":"mistral-medium","parameters":128,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"mistralai/mistral-nemotron","name":"mistral-nemotron","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-nemo","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-24b-instruct","name":"mistral-small-24b-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":24,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-3.1-24b-instruct-2503","name":"mistral-small-3.1-24b-instruct-2503","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","alias":"mistralai/mistral-small-3.1-24b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":24,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-3.1-24b-instruct","name":"mistral-small-3.1-24b-instruct-2503","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/mistral-small-3.1-24b-instruct-2503"],"model_type":"chat","parameters":24,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-4-119b-2603","name":"mistral-small-4-119b-2603","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","alias":"mistralai/mistral-small-4-119b","capabilities":{"streaming":true},"model_type":"chat","parameters":119,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-4-119b","name":"mistral-small-4-119b-2603","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","capabilities":{"streaming":true},"snapshots":["mistralai/mistral-small-4-119b-2603"],"model_type":"chat","parameters":119,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mixtral-8x22b-instruct-v0.1","name":"mixtral-8x22b-instruct-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mixtral-8x22b-v0.1","name":"mixtral-8x22b-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mixtral-8x7b-instruct-v0.1","name":"mixtral-8x7b-instruct-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2-instruct-0905","name":"kimi-k2-instruct-0905","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2-instruct","name":"kimi-k2-instruct","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2-thinking","name":"kimi-k2-thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"family":"kimi-k2","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Long-term thinking, multi-step tool usage, complex problem solving.","tagline":"Long-term thinking, multi-step tool usage, complex problem solving.","context_window":256000,"max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.5","name":"kimi-k2.5","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"family":"kimi-k2.5","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","tagline":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","context_window":256000,"max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.6","name":"kimi-k2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"nv-mistralai/mistral-nemo-12b-instruct","name":"mistral-nemo-12b-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-nemo","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/ai-synthetic-video-detector","name":"ai-synthetic-video-detector","created_by":"nvidia","source":"official","capabilities":{"streaming":true},"last_updated":"2026-04-24"},{"id":"nvidia/cosmos-reason2-8b","name":"cosmos-reason2-8b","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"reasoning":true},"parameters":8},{"id":"nvidia/embed-qa-4","name":"embed-qa-4","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed","family":"embed"},{"id":"nvidia/gliner-pii","name":"gliner-pii","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/ising-calibration-1-35b-a3b","name":"ising-calibration-1-35b-a3b","created_by":"nvidia","source":"official","parameters":35,"active_parameters":3,"capabilities":{"streaming":true},"last_updated":"2026-04-16"},{"id":"nvidia/llama-3.1-nemoguard-8b-content-safety","name":"llama-3.1-nemoguard-8b-content-safety","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true},"model_type":"moderation","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemoguard-8b-topic-control","name":"llama-3.1-nemoguard-8b-topic-control","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true},"model_type":"moderation","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-51b-instruct","name":"llama-3.1-nemotron-51b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":51,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-70b-instruct","name":"llama-3.1-nemotron-70b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-70b-reward","name":"llama-3.1-nemotron-70b-reward","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-nano-4b-v1.1","name":"llama-3.1-nemotron-nano-4b-v1.1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-nano-8b-v1","name":"llama-3.1-nemotron-nano-8b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-nano-vl-8b-v1","name":"llama-3.1-nemotron-nano-vl-8b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-safety-guard-8b-v3","name":"llama-3.1-nemotron-safety-guard-8b-v3","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"moderation","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-ultra-253b-v1","name":"llama-3.1-nemotron-ultra-253b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":253,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.2-nemoretriever-1b-vlm-embed-v1","name":"llama-3.2-nemoretriever-1b-vlm-embed-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true,"vision":true},"model_type":"embed","parameters":1},{"id":"nvidia/llama-3.2-nemoretriever-300m-embed-v1","name":"llama-3.2-nemoretriever-300m-embed-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true},"model_type":"embed"},{"id":"nvidia/llama-3.2-nv-embedqa-1b-v1","name":"llama-3.2-nv-embedqa-1b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true},"model_type":"embed","parameters":1},{"id":"nvidia/llama-3.2-nv-embedqa-1b-v2","name":"llama-3.2-nv-embedqa-1b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true},"model_type":"embed","parameters":1},{"id":"nvidia/llama-3.3-nemotron-super-49b-v1.5","name":"llama-3.3-nemotron-super-49b-v1.5","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":49,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.3-nemotron-super-49b-v1","name":"llama-3.3-nemotron-super-49b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":49,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-nemotron-embed-1b-v2","name":"llama-nemotron-embed-1b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"embed","parameters":1,"tools":["function_calling"]},{"id":"nvidia/llama-nemotron-embed-vl-1b-v2","name":"llama-nemotron-embed-vl-1b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"embed","parameters":1,"tools":["function_calling"]},{"id":"nvidia/llama3-chatqa-1.5-70b","name":"llama3-chatqa-1.5-70b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama3-chatqa-1.5-8b","name":"llama3-chatqa-1.5-8b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/mistral-nemo-minitron-8b-8k-instruct","name":"mistral-nemo-minitron-8b-8k-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-nemo","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/mistral-nemo-minitron-8b-base","name":"mistral-nemo-minitron-8b-base","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","family":"mistral-nemo","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemoretriever-parse","name":"nemoretriever-parse","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/nemotron-3-content-safety","name":"nemotron-3-content-safety","created_by":"nvidia","source":"official","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"nemotron-3-nano-30b-a3b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":30,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning","name":"nemotron-3-nano-omni-30b-a3b-reasoning","created_by":"nvidia","source":"official","parameters":30,"active_parameters":3,"model_type":"reasoning","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"nemotron-3-super-120b-a12b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":120,"active_parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"nemotron-3-ultra-550b-a55b","created_by":"nvidia","source":"official","parameters":550,"active_parameters":55,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-3.5-content-safety","name":"nemotron-3.5-content-safety","created_by":"nvidia","source":"official","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"nvidia/nemotron-4-340b-instruct","name":"nemotron-4-340b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":340,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-4-340b-reward","name":"nemotron-4-340b-reward","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":340,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-4-mini-hindi-4b-instruct","name":"nemotron-4-mini-hindi-4b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-content-safety-reasoning-4b","name":"nemotron-content-safety-reasoning-4b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"model_type":"reasoning","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-mini-4b-instruct","name":"nemotron-mini-4b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-nano-12b-v2-vl","name":"nemotron-nano-12b-v2-vl","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"chat","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-nano-3-30b-a3b","name":"nemotron-nano-3-30b-a3b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":30,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-parse","name":"nemotron-parse","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/neva-22b","name":"neva-22b","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":22},{"id":"nvidia/nv-embed-v1","name":"nv-embed-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed"},{"id":"nvidia/nv-embedcode-7b-v1","name":"nv-embedcode-7b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed","parameters":7},{"id":"nvidia/nv-embedqa-e5-v5","name":"nv-embedqa-e5-v5","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed"},{"id":"nvidia/nv-embedqa-mistral-7b-v2","name":"nv-embedqa-mistral-7b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed","parameters":7},{"id":"nvidia/nvclip","name":"nvclip","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/nvidia-nemotron-nano-9b-v2","name":"nvidia-nemotron-nano-9b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":9,"tools":["function_calling"]},{"id":"nvidia/riva-translate-4b-instruct-v1.1","name":"riva-translate-4b-instruct-v1.1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"translation","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/riva-translate-4b-instruct","name":"riva-translate-4b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"translation","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/streampetr","name":"streampetr","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/usdcode-llama-3.1-70b-instruct","name":"usdcode-llama-3.1-70b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":70,"tools":["function_calling"]},{"id":"nvidia/vila","name":"vila","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":120,"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Most powerful open-weight model, fits into an H100 GPU","context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":20,"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Medium-sized open-weight model for low latency","context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"opengpt-x/teuken-7b-instruct-commercial-v0.4","name":"teuken-7b-instruct-commercial-v0.4","created_by":"opengpt-x","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"qwen/qwen2-7b-instruct","name":"qwen2-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","context_window":131072,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen2.5-7b-instruct","name":"qwen2.5-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2.5","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","context_window":1000000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen2.5-coder-32b-instruct","name":"qwen2.5-coder-32b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2.5","capabilities":{"streaming":true,"tool_call":true},"model_type":"code","parameters":32,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen/qwen2.5-coder-7b-instruct","name":"qwen2.5-coder-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2.5","capabilities":{"streaming":true,"tool_call":true},"model_type":"code","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen/qwen3-coder-480b-a35b-instruct","name":"qwen3-coder-480b-a35b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","parameters":480,"active_parameters":35,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":256000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"qwen3-next-80b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","parameters":80,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":128000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"qwen3-next-80b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","parameters":80,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"context_window":128000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-122b-a10b","name":"qwen3.5-122b-a10b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":122,"active_parameters":10,"modalities":{"input":["text"],"output":["text"]},"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","context_window":256000,"max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-397b-a17b","name":"qwen3.5-397b-a17b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":397,"active_parameters":17,"modalities":{"input":["text"],"output":["text"]},"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","context_window":256000,"max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwq-32b","name":"qwq-32b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwq","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","parameters":32,"modalities":{"input":["text"],"output":["text"]},"description":"QwQ is a reasoning model trained from Qwen2.5-32B, with significantly enhanced reasoning capabilities through reinforcement learning. Its core metrics (AIME 24/25, LiveCodeBench) and some general metrics (IFEval, LiveBench, etc.) match DeepSeek-R1 full version levels and significantly outperform DeepSeek-R1-Distill-Qwen-32B, which is also based on Qwen2.5-32B. [Usage](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"QwQ is an inference model trained on Qwen2.5-32B.","context_window":131072,"max_output_tokens":32768,"license":"apache-2.0","open_weight":true},{"id":"rakuten/rakutenai-7b-chat","name":"rakutenai-7b-chat","created_by":"rakuten","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"rakuten/rakutenai-7b-instruct","name":"rakutenai-7b-instruct","created_by":"rakuten","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"sarvamai/sarvam-m","name":"sarvam-m","created_by":"sarvamai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"snowflake/arctic-embed-l","name":"arctic-embed-l","created_by":"snowflake","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed"},{"id":"speakleash/bielik-11b-v2.3-instruct","name":"bielik-11b-v2.3-instruct","created_by":"speakleash","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":11,"tools":["function_calling"]},{"id":"speakleash/bielik-11b-v2.6-instruct","name":"bielik-11b-v2.6-instruct","created_by":"speakleash","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":11,"tools":["function_calling"]},{"id":"stepfun-ai/step-3.5-flash","name":"step-3.5-flash","created_by":"stepfun","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":true},"description":"Flagship reasoning model with 256k context, MoE architecture (199B total, ~11B active). Optimized for agentic tasks, tool calling, planning, math, and coding.","tagline":"Frontier reasoning and agentic capabilities at low cost.","context_window":256000,"license":"apache-2.0","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"stepfun-ai/step-3.7-flash","name":"step-3.7-flash","created_by":"stepfun","source":"official","capabilities":{"streaming":true},"last_updated":"2026-06-02"},{"id":"stockmark/stockmark-2-100b-instruct","name":"stockmark-2-100b-instruct","created_by":"stockmark","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":100,"tools":["function_calling"]},{"id":"thudm/chatglm3-6b","name":"chatglm3-6b","created_by":"zhipu","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":6,"tools":["function_calling"]},{"id":"tiiuae/falcon3-7b-instruct","name":"falcon3-7b-instruct","created_by":"tii","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"tokyotech-llm/llama-3-swallow-70b-instruct-v0.1","name":"llama-3-swallow-70b-instruct-v0.1","created_by":"tokyotech-llm","source":"official","last_updated":"2026-03-23","family":"llama-3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"upstage/solar-10.7b-instruct","name":"solar-10.7b-instruct","created_by":"upstage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":10.7,"tools":["function_calling"]},{"id":"utter-project/eurollm-9b-instruct","name":"eurollm-9b-instruct","created_by":"utter-project","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":9,"tools":["function_calling"]},{"id":"writer/palmyra-creative-122b","name":"palmyra-creative-122b","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":122,"modalities":{"input":["text"],"output":["text"]}},{"id":"writer/palmyra-fin-70b-32k","name":"palmyra-fin-70b-32k","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"writer/palmyra-med-70b-32k","name":"palmyra-med-70b-32k","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"writer/palmyra-med-70b","name":"palmyra-med-70b","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"yentinglin/llama-3-taiwan-70b-instruct","name":"llama-3-taiwan-70b-instruct","created_by":"yentinglin","source":"official","last_updated":"2026-03-23","family":"llama-3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"z-ai/glm-5.1","name":"glm-5.1","created_by":"zhipu","source":"official","family":"glm-5.1","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21"},{"id":"z-ai/glm4.7","name":"glm4.7","created_by":"zhipu","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","family":"glm","modalities":{"input":["text"],"output":["text"]}},{"id":"z-ai/glm5","name":"glm5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","family":"glm","modalities":{"input":["text"],"output":["text"]}},{"id":"zyphra/zamba2-7b-instruct","name":"zamba2-7b-instruct","created_by":"zyphra","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]}]};
+export const provider: ProviderWithModels = {"id":"nvidia","name":"NVIDIA","region":"US","headquarters":"Santa Clara, CA","founded":1993,"url":"https://build.nvidia.com","api_url":"https://integrate.api.nvidia.com/v1","docs_url":"https://docs.api.nvidia.com","pricing_url":"https://build.nvidia.com/pricing","description":"NVIDIA NIM provides optimized inference for 190+ models on NVIDIA GPU infrastructure.","type":"cloud","playground_url":"https://build.nvidia.com","sdk":{"python":"openai"},"openai_compatible":true,"free_tier":true,"models_url":"https://build.nvidia.com/explore/discover","twitter_url":"https://x.com/nvidia","blog_url":"https://blogs.nvidia.com","terms_url":"https://developer.nvidia.com/legal/terms","support_url":"https://forums.developer.nvidia.com","icon":"","models":[{"id":"01-ai/yi-large","name":"yi-large","created_by":"01-ai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"abacusai/dracarys-llama-3.1-70b-instruct","name":"dracarys-llama-3.1-70b-instruct","created_by":"abacusai","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":70,"tools":["function_calling"]},{"id":"adept/fuyu-8b","name":"fuyu-8b","created_by":"adept","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":8},{"id":"ai21labs/jamba-1.5-large-instruct","name":"jamba-1.5-large-instruct","created_by":"ai21","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"ai21labs/jamba-1.5-mini-instruct","name":"jamba-1.5-mini-instruct","created_by":"ai21","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"aisingapore/sea-lion-7b-instruct","name":"sea-lion-7b-instruct","created_by":"aisingapore","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"baai/bge-m3","name":"bge-m3","created_by":"baai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"baichuan-inc/baichuan2-13b-chat","name":"baichuan2-13b-chat","created_by":"baichuan","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":13,"tools":["function_calling"]},{"id":"bigcode/starcoder2-15b","name":"starcoder2-15b","created_by":"bigcode","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":15},{"id":"bigcode/starcoder2-7b","name":"starcoder2-7b","created_by":"bigcode","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"bytedance/seed-oss-36b-instruct","name":"seed-oss-36b-instruct","created_by":"bytedance","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":36,"tools":["function_calling"]},{"id":"databricks/dbrx-instruct","name":"dbrx-instruct","created_by":"databricks","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"]},{"id":"deepseek-ai/deepseek-coder-6.7b-instruct","name":"deepseek-coder-6.7b-instruct","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true,"tool_call":true},"parameters":6.7,"tools":["function_calling"]},{"id":"deepseek-ai/deepseek-r1-distill-llama-8b","name":"deepseek-r1-distill-llama-8b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-r1-distill-qwen-14b","name":"deepseek-r1-distill-qwen-14b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":14,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-r1-distill-qwen-32b","name":"deepseek-r1-distill-qwen-32b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":32,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-r1-distill-qwen-7b","name":"deepseek-r1-distill-qwen-7b","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek-r1","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v3.1-terminus","name":"deepseek-v3.1-terminus","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v3.1","name":"deepseek-v3.1","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v3.2","name":"deepseek-v3.2","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-ai/deepseek-v4-flash","name":"deepseek-v4-flash","created_by":"deepseek","source":"official","family":"deepseek","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","description":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","context_window":1000000,"max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"deepseek-ai/deepseek-v4-pro","name":"deepseek-v4-pro","created_by":"deepseek","source":"official","family":"deepseek","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","description":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","context_window":1000000,"max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"google/codegemma-1.1-7b","name":"codegemma-1.1-7b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"google/codegemma-7b","name":"codegemma-7b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":7},{"id":"google/deplot","name":"deplot","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"google/diffusiongemma-26b-a4b-it","name":"diffusiongemma-26b-a4b-it","created_by":"google","source":"official","parameters":26,"active_parameters":4,"capabilities":{"streaming":true},"last_updated":"2026-06-18"},{"id":"google/gemma-2-27b-it","name":"gemma-2-27b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true},"model_type":"chat","parameters":27,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2-2b-it","name":"gemma-2-2b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true},"model_type":"chat","parameters":2,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2-9b-it","name":"gemma-2-9b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true},"model_type":"chat","parameters":9,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-2b","name":"gemma-2b","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2b","capabilities":{"streaming":true},"model_type":"chat","parameters":2,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-12b-it","name":"gemma-3-12b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":12,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-1b-it","name":"gemma-3-1b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":1,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-27b-it","name":"gemma-3-27b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":27,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3-4b-it","name":"gemma-3-4b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","capabilities":{"streaming":true},"model_type":"chat","parameters":4,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3n-e2b-it","name":"gemma-3n-e2b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3n","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-3n-e4b-it","name":"gemma-3n-e4b-it","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3n","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-4-31b-it","name":"gemma-4-31b-it","created_by":"google","source":"official","family":"gemma-4","parameters":31,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-03"},{"id":"google/gemma-7b","name":"gemma-7b","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-7b","capabilities":{"streaming":true},"model_type":"chat","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/paligemma","name":"paligemma","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"google/recurrentgemma-2b","name":"recurrentgemma-2b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":2},{"id":"google/shieldgemma-9b","name":"shieldgemma-9b","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":9},{"id":"gotocompany/gemma-2-9b-cpt-sahabatai-instruct","name":"gemma-2-9b-cpt-sahabatai-instruct","created_by":"gotocompany","source":"official","last_updated":"2026-03-23","family":"gemma-2","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":9,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-3.0-3b-a800m-instruct","name":"granite-3.0-3b-a800m-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":3,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-3.0-8b-instruct","name":"granite-3.0-8b-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-3.3-8b-instruct","name":"granite-3.3-8b-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-34b-code-instruct","name":"granite-34b-code-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":34,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-8b-code-instruct","name":"granite-8b-code-instruct","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"ibm/granite-guardian-3.0-8b","name":"granite-guardian-3.0-8b","created_by":"ibm","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"moderation","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"igenius/colosseum_355b_instruct_16k","name":"colosseum_355b_instruct_16k","created_by":"igenius","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":355,"tools":["function_calling"]},{"id":"igenius/italia_10b_instruct_16k","name":"italia_10b_instruct_16k","created_by":"igenius","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":10,"tools":["function_calling"]},{"id":"institute-of-science-tokyo/llama-3.1-swallow-70b-instruct-v0.1","name":"llama-3.1-swallow-70b-instruct-v0.1","created_by":"institute-of-science-tokyo","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"institute-of-science-tokyo/llama-3.1-swallow-8b-instruct-v0.1","name":"llama-3.1-swallow-8b-instruct-v0.1","created_by":"institute-of-science-tokyo","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"marin/marin-8b-instruct","name":"marin-8b-instruct","created_by":"marin","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":8,"tools":["function_calling"]},{"id":"mediatek/breeze-7b-instruct","name":"breeze-7b-instruct","created_by":"mediatek","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"meta/codellama-70b","name":"codellama-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"family":"codellama","model_type":"code","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.1-405b-instruct","name":"llama-3.1-405b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":405,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta/llama-3.1-70b-instruct","name":"llama-3.1-70b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta/llama-3.1-8b-instruct","name":"llama-3.1-8b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.1","open_weight":true},{"id":"meta/llama-3.2-11b-vision-instruct","name":"llama-3.2-11b-vision-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":11,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.2-1b-instruct","name":"llama-3.2-1b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":1,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.2-3b-instruct","name":"llama-3.2-3b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.2-90b-vision-instruct","name":"llama-3.2-90b-vision-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":90,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.2","open_weight":true},{"id":"meta/llama-3.3-70b-instruct","name":"llama-3.3-70b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2023-12","license":"llama3.3","open_weight":true},{"id":"meta/llama-4-maverick-17b-128e-instruct","name":"llama-4-maverick-17b-128e-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":17,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":16384,"license":"llama4","open_weight":true},{"id":"meta/llama-4-scout-17b-16e-instruct","name":"llama-4-scout-17b-16e-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"model_type":"chat","parameters":17,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":16384,"license":"llama4","open_weight":true},{"id":"meta/llama-guard-4-12b","name":"llama-guard-4-12b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-guard","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"moderation","parameters":12,"modalities":{"input":["text"],"output":["text"]},"license":"llama4","open_weight":true},{"id":"meta/llama2-70b","name":"llama2-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama3-70b-instruct","name":"llama3-70b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama3-8b-instruct","name":"llama3-8b-instruct","created_by":"meta","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/kosmos-2","name":"kosmos-2","created_by":"microsoft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"microsoft/phi-3-medium-128k-instruct","name":"phi-3-medium-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-medium-4k-instruct","name":"phi-3-medium-4k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-mini-128k-instruct","name":"phi-3-mini-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-mini-4k-instruct","name":"phi-3-mini-4k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-small-128k-instruct","name":"phi-3-small-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-small-8k-instruct","name":"phi-3-small-8k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3-vision-128k-instruct","name":"phi-3-vision-128k-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3.5-mini-instruct","name":"phi-3.5-mini-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3.5-moe-instruct","name":"phi-3.5-moe-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-3.5-vision-instruct","name":"phi-3.5-vision-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-4-mini-flash-reasoning","name":"phi-4-mini-flash-reasoning","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"reasoning":true},"model_type":"reasoning","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-4-mini-instruct","name":"phi-4-mini-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"microsoft/phi-4-multimodal-instruct","name":"phi-4-multimodal-instruct","created_by":"microsoft","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"minimaxai/minimax-m2.1","name":"minimax-m2.1","created_by":"minimax","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","tagline":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","context_window":204800,"license":"mit","open_weight":true},{"id":"minimaxai/minimax-m2.5","name":"minimax-m2.5","created_by":"minimax","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Peak Performance. Ultimate Value. Master the Complex","tagline":"Peak Performance.","context_window":204800,"license":"mit","open_weight":true},{"id":"minimaxai/minimax-m2.7","name":"minimax-m2.7","created_by":"minimax","source":"official","family":"minimax","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-12","description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","context_window":204800,"license":"proprietary","open_weight":false},{"id":"minimaxai/minimax-m3","name":"minimax-m3","created_by":"minimax","source":"official","family":"minimax","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","context_window":1000000,"license":"proprietary","open_weight":false},{"id":"mistralai/codestral-22b-instruct-v0.1","name":"codestral-22b-instruct-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"codestral","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":22,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/devstral-2-123b-instruct-2512","name":"devstral-2-123b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","alias":"mistralai/devstral-2-123b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":123,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/devstral-2-123b-instruct","name":"devstral-2-123b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/devstral-2-123b-instruct-2512"],"model_type":"chat","parameters":123,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/magistral-small-2506","name":"magistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-03-23","alias":"mistralai/magistral-small","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"description":"Our small reasoning model released July 2025.","tagline":"Our small reasoning model released July 2025.","context_window":40000,"license":"apache-2.0","open_weight":true},{"id":"mistralai/magistral-small","name":"magistral-small-2506","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"snapshots":["mistralai/magistral-small-2506"],"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"description":"Our small reasoning model released July 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","context_window":40000,"license":"apache-2.0","open_weight":true},{"id":"mistralai/mamba-codestral-7b-v0.1","name":"mamba-codestral-7b-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mathstral-7b-v0.1","name":"mathstral-7b-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":7,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/ministral-14b-instruct-2512","name":"ministral-14b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","alias":"mistralai/ministral-14b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":14,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/ministral-14b-instruct","name":"ministral-14b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/ministral-14b-instruct-2512"],"model_type":"chat","parameters":14,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-7b-instruct-v0.2","name":"mistral-7b-instruct-v0.2","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-7b","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-7b-instruct-v0.3","name":"mistral-7b-instruct-v0.3","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-7b","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large-2-instruct","name":"mistral-large-2-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large-3-675b-instruct-2512","name":"mistral-large-3-675b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","alias":"mistralai/mistral-large-3-675b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":675,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large-3-675b-instruct","name":"mistral-large-3-675b-instruct-2512","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/mistral-large-3-675b-instruct-2512"],"model_type":"chat","parameters":675,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-large","name":"mistral-large","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Our first and larger SOTA model for enterprise use cases with powerfull chat capabilities.","tagline":"Our top-tier large model for high-complexity tasks with the lastest version released November 2024.","context_window":32000,"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-medium-3-instruct","name":"mistral-medium-3-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-medium","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-medium-3.5-128b","name":"mistral-medium-3.5-128b","created_by":"mistral","source":"official","family":"mistral-medium","parameters":128,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"mistralai/mistral-nemotron","name":"mistral-nemotron","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-nemo","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-24b-instruct","name":"mistral-small-24b-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":24,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-3.1-24b-instruct-2503","name":"mistral-small-3.1-24b-instruct-2503","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","alias":"mistralai/mistral-small-3.1-24b-instruct","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":24,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-3.1-24b-instruct","name":"mistral-small-3.1-24b-instruct-2503","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","capabilities":{"streaming":true,"tool_call":true},"snapshots":["mistralai/mistral-small-3.1-24b-instruct-2503"],"model_type":"chat","parameters":24,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-4-119b-2603","name":"mistral-small-4-119b-2603","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","alias":"mistralai/mistral-small-4-119b","capabilities":{"streaming":true},"model_type":"chat","parameters":119,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mistral-small-4-119b","name":"mistral-small-4-119b-2603","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","capabilities":{"streaming":true},"snapshots":["mistralai/mistral-small-4-119b-2603"],"model_type":"chat","parameters":119,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mixtral-8x22b-instruct-v0.1","name":"mixtral-8x22b-instruct-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mixtral-8x22b-v0.1","name":"mixtral-8x22b-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"mistralai/mixtral-8x7b-instruct-v0.1","name":"mixtral-8x7b-instruct-v0.1","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2-instruct-0905","name":"kimi-k2-instruct-0905","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2-instruct","name":"kimi-k2-instruct","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2-thinking","name":"kimi-k2-thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"family":"kimi-k2","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Long-term thinking, multi-step tool usage, complex problem solving.","tagline":"Long-term thinking, multi-step tool usage, complex problem solving.","context_window":256000,"max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.5","name":"kimi-k2.5","created_by":"moonshot","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"family":"kimi-k2.5","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","tagline":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","context_window":256000,"max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.6","name":"kimi-k2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"nv-mistralai/mistral-nemo-12b-instruct","name":"mistral-nemo-12b-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-nemo","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/ai-synthetic-video-detector","name":"ai-synthetic-video-detector","created_by":"nvidia","source":"official","capabilities":{"streaming":true},"last_updated":"2026-04-24"},{"id":"nvidia/cosmos-reason2-8b","name":"cosmos-reason2-8b","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"reasoning":true},"parameters":8},{"id":"nvidia/embed-qa-4","name":"embed-qa-4","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed","family":"embed"},{"id":"nvidia/gliner-pii","name":"gliner-pii","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/ising-calibration-1-35b-a3b","name":"ising-calibration-1-35b-a3b","created_by":"nvidia","source":"official","parameters":35,"active_parameters":3,"capabilities":{"streaming":true},"last_updated":"2026-04-16"},{"id":"nvidia/llama-3.1-nemoguard-8b-content-safety","name":"llama-3.1-nemoguard-8b-content-safety","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true},"model_type":"moderation","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemoguard-8b-topic-control","name":"llama-3.1-nemoguard-8b-topic-control","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true},"model_type":"moderation","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-51b-instruct","name":"llama-3.1-nemotron-51b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":51,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-70b-instruct","name":"llama-3.1-nemotron-70b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-70b-reward","name":"llama-3.1-nemotron-70b-reward","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-nano-4b-v1.1","name":"llama-3.1-nemotron-nano-4b-v1.1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-nano-8b-v1","name":"llama-3.1-nemotron-nano-8b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-nano-vl-8b-v1","name":"llama-3.1-nemotron-nano-vl-8b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-safety-guard-8b-v3","name":"llama-3.1-nemotron-safety-guard-8b-v3","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"moderation","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.1-nemotron-ultra-253b-v1","name":"llama-3.1-nemotron-ultra-253b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.1","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":253,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.2-nemoretriever-1b-vlm-embed-v1","name":"llama-3.2-nemoretriever-1b-vlm-embed-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true,"vision":true},"model_type":"embed","parameters":1},{"id":"nvidia/llama-3.2-nemoretriever-300m-embed-v1","name":"llama-3.2-nemoretriever-300m-embed-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true},"model_type":"embed"},{"id":"nvidia/llama-3.2-nv-embedqa-1b-v1","name":"llama-3.2-nv-embedqa-1b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true},"model_type":"embed","parameters":1},{"id":"nvidia/llama-3.2-nv-embedqa-1b-v2","name":"llama-3.2-nv-embedqa-1b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-21","family":"llama-3.2","capabilities":{"streaming":true},"model_type":"embed","parameters":1},{"id":"nvidia/llama-3.3-nemotron-super-49b-v1.5","name":"llama-3.3-nemotron-super-49b-v1.5","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":49,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-3.3-nemotron-super-49b-v1","name":"llama-3.3-nemotron-super-49b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","family":"llama-3.3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":49,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama-nemotron-embed-1b-v2","name":"llama-nemotron-embed-1b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"embed","parameters":1,"tools":["function_calling"]},{"id":"nvidia/llama-nemotron-embed-vl-1b-v2","name":"llama-nemotron-embed-vl-1b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"embed","parameters":1,"tools":["function_calling"]},{"id":"nvidia/llama3-chatqa-1.5-70b","name":"llama3-chatqa-1.5-70b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/llama3-chatqa-1.5-8b","name":"llama3-chatqa-1.5-8b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/mistral-nemo-minitron-8b-8k-instruct","name":"mistral-nemo-minitron-8b-8k-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","family":"mistral-nemo","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/mistral-nemo-minitron-8b-base","name":"mistral-nemo-minitron-8b-base","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","family":"mistral-nemo","parameters":8,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemoretriever-parse","name":"nemoretriever-parse","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/nemotron-3-content-safety","name":"nemotron-3-content-safety","created_by":"nvidia","source":"official","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"nemotron-3-nano-30b-a3b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":30,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning","name":"nemotron-3-nano-omni-30b-a3b-reasoning","created_by":"nvidia","source":"official","parameters":30,"active_parameters":3,"model_type":"reasoning","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"nemotron-3-super-120b-a12b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":120,"active_parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"nemotron-3-ultra-550b-a55b","created_by":"nvidia","source":"official","parameters":550,"active_parameters":55,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-3.5-content-safety","name":"nemotron-3.5-content-safety","created_by":"nvidia","source":"official","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"nvidia/nemotron-4-340b-instruct","name":"nemotron-4-340b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":340,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-4-340b-reward","name":"nemotron-4-340b-reward","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":340,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-4-mini-hindi-4b-instruct","name":"nemotron-4-mini-hindi-4b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-content-safety-reasoning-4b","name":"nemotron-content-safety-reasoning-4b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"model_type":"reasoning","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-mini-4b-instruct","name":"nemotron-mini-4b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-nano-12b-v2-vl","name":"nemotron-nano-12b-v2-vl","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"chat","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-nano-3-30b-a3b","name":"nemotron-nano-3-30b-a3b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":30,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-parse","name":"nemotron-parse","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/neva-22b","name":"neva-22b","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"parameters":22},{"id":"nvidia/nv-embed-v1","name":"nv-embed-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed"},{"id":"nvidia/nv-embedcode-7b-v1","name":"nv-embedcode-7b-v1","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed","parameters":7},{"id":"nvidia/nv-embedqa-e5-v5","name":"nv-embedqa-e5-v5","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed"},{"id":"nvidia/nv-embedqa-mistral-7b-v2","name":"nv-embedqa-mistral-7b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed","parameters":7},{"id":"nvidia/nvclip","name":"nvclip","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/nvidia-nemotron-nano-9b-v2","name":"nvidia-nemotron-nano-9b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":9,"tools":["function_calling"]},{"id":"nvidia/riva-translate-4b-instruct-v1.1","name":"riva-translate-4b-instruct-v1.1","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"translation","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/riva-translate-4b-instruct","name":"riva-translate-4b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"model_type":"translation","parameters":4,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/streampetr","name":"streampetr","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"nvidia/usdcode-llama-3.1-70b-instruct","name":"usdcode-llama-3.1-70b-instruct","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":70,"tools":["function_calling"]},{"id":"nvidia/vila","name":"vila","created_by":"nvidia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":120,"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Most powerful open-weight model, fits into an H100 GPU","context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"model_type":"chat","parameters":20,"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Medium-sized open-weight model for low latency","context_window":131072,"max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"opengpt-x/teuken-7b-instruct-commercial-v0.4","name":"teuken-7b-instruct-commercial-v0.4","created_by":"opengpt-x","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"qwen/qwen2-7b-instruct","name":"qwen2-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","context_window":131072,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen2.5-7b-instruct","name":"qwen2.5-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2.5","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","context_window":1000000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen2.5-coder-32b-instruct","name":"qwen2.5-coder-32b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2.5","capabilities":{"streaming":true,"tool_call":true},"model_type":"code","parameters":32,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen/qwen2.5-coder-7b-instruct","name":"qwen2.5-coder-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen2.5","capabilities":{"streaming":true,"tool_call":true},"model_type":"code","parameters":7,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen/qwen3-coder-480b-a35b-instruct","name":"qwen3-coder-480b-a35b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","parameters":480,"active_parameters":35,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":256000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"qwen3-next-80b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","parameters":80,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"context_window":256000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"qwen3-next-80b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","parameters":80,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"context_window":256000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-122b-a10b","name":"qwen3.5-122b-a10b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":122,"active_parameters":10,"modalities":{"input":["text"],"output":["text"]},"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","context_window":256000,"max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-397b-a17b","name":"qwen3.5-397b-a17b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","parameters":397,"active_parameters":17,"modalities":{"input":["text"],"output":["text"]},"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","context_window":256000,"max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwq-32b","name":"qwq-32b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwq","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","parameters":32,"modalities":{"input":["text"],"output":["text"]},"description":"QwQ is a reasoning model trained from Qwen2.5-32B, with significantly enhanced reasoning capabilities through reinforcement learning. Its core metrics (AIME 24/25, LiveCodeBench) and some general metrics (IFEval, LiveBench, etc.) match DeepSeek-R1 full version levels and significantly outperform DeepSeek-R1-Distill-Qwen-32B, which is also based on Qwen2.5-32B. [Usage](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"QwQ is an inference model trained on Qwen2.5-32B.","context_window":131072,"max_output_tokens":32768,"license":"apache-2.0","open_weight":true},{"id":"rakuten/rakutenai-7b-chat","name":"rakutenai-7b-chat","created_by":"rakuten","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"rakuten/rakutenai-7b-instruct","name":"rakutenai-7b-instruct","created_by":"rakuten","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]},{"id":"sarvamai/sarvam-m","name":"sarvam-m","created_by":"sarvamai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true}},{"id":"snowflake/arctic-embed-l","name":"arctic-embed-l","created_by":"snowflake","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"model_type":"embed"},{"id":"speakleash/bielik-11b-v2.3-instruct","name":"bielik-11b-v2.3-instruct","created_by":"speakleash","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":11,"tools":["function_calling"]},{"id":"speakleash/bielik-11b-v2.6-instruct","name":"bielik-11b-v2.6-instruct","created_by":"speakleash","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":11,"tools":["function_calling"]},{"id":"stepfun-ai/step-3.5-flash","name":"step-3.5-flash","created_by":"stepfun","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":true},"description":"Flagship reasoning model with 256k context, MoE architecture (199B total, ~11B active). Optimized for agentic tasks, tool calling, planning, math, and coding.","tagline":"Frontier reasoning and agentic capabilities at low cost.","context_window":256000,"license":"apache-2.0","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"stepfun-ai/step-3.7-flash","name":"step-3.7-flash","created_by":"stepfun","source":"official","capabilities":{"streaming":true},"last_updated":"2026-06-02"},{"id":"stockmark/stockmark-2-100b-instruct","name":"stockmark-2-100b-instruct","created_by":"stockmark","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":100,"tools":["function_calling"]},{"id":"thudm/chatglm3-6b","name":"chatglm3-6b","created_by":"zhipu","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":6,"tools":["function_calling"]},{"id":"tiiuae/falcon3-7b-instruct","name":"falcon3-7b-instruct","created_by":"tii","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"tokyotech-llm/llama-3-swallow-70b-instruct-v0.1","name":"llama-3-swallow-70b-instruct-v0.1","created_by":"tokyotech-llm","source":"official","last_updated":"2026-03-23","family":"llama-3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"upstage/solar-10.7b-instruct","name":"solar-10.7b-instruct","created_by":"upstage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":10.7,"tools":["function_calling"]},{"id":"utter-project/eurollm-9b-instruct","name":"eurollm-9b-instruct","created_by":"utter-project","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":9,"tools":["function_calling"]},{"id":"writer/palmyra-creative-122b","name":"palmyra-creative-122b","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":122,"modalities":{"input":["text"],"output":["text"]}},{"id":"writer/palmyra-fin-70b-32k","name":"palmyra-fin-70b-32k","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"writer/palmyra-med-70b-32k","name":"palmyra-med-70b-32k","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"writer/palmyra-med-70b","name":"palmyra-med-70b","created_by":"writer","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","parameters":70,"modalities":{"input":["text"],"output":["text"]}},{"id":"yentinglin/llama-3-taiwan-70b-instruct","name":"llama-3-taiwan-70b-instruct","created_by":"yentinglin","source":"official","last_updated":"2026-03-23","family":"llama-3","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"z-ai/glm-5.1","name":"glm-5.1","created_by":"zhipu","source":"official","family":"glm-5.1","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21"},{"id":"z-ai/glm4.7","name":"glm4.7","created_by":"zhipu","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","family":"glm","modalities":{"input":["text"],"output":["text"]}},{"id":"z-ai/glm5","name":"glm5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"model_type":"chat","family":"glm","modalities":{"input":["text"],"output":["text"]}},{"id":"zyphra/zamba2-7b-instruct","name":"zamba2-7b-instruct","created_by":"zyphra","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"parameters":7,"tools":["function_calling"]}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/ollama.ts b/packages/npm/src/providers/ollama.ts
index d7e02b0ef..c4ab06cb3 100644
--- a/packages/npm/src/providers/ollama.ts
+++ b/packages/npm/src/providers/ollama.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"ollama","name":"Ollama","region":"US","headquarters":"Palo Alto, CA","founded":2023,"url":"https://ollama.com","api_url":"http://localhost:11434/api","docs_url":"https://github.com/ollama/ollama/blob/main/docs/api.md","pricing_url":"https://ollama.com","description":"Run open-source LLMs locally on your machine with a simple CLI.","type":"aggregator","openai_compatible":true,"free_tier":true,"github_url":"https://github.com/ollama/ollama","models_url":"https://ollama.com/library","twitter_url":"https://x.com/ollama","discord_url":"https://discord.com/invite/ollama","blog_url":"https://ollama.com/blog","support_url":"https://github.com/ollama/ollama","icon":"","models":[{"id":"cogito-2.1","name":"cogito-2.1","created_by":"deepcogito","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"description":"The Cogito v2.1 LLMs are instruction tuned generative models. All models are released under MIT license for commercial use.","release_date":"2025-11-19","parameters":671,"page_url":"https://ollama.com/library/cogito-2.1","tagline":"The Cogito v2.1 LLMs are instruction tuned generative models.","context_window":163840,"license":"mit","open_weight":true,"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-v3.1","name":"deepseek-v3.1","created_by":"deepseek","source":"official","last_updated":"2026-03-24","family":"deepseek","capabilities":{"streaming":true,"fine_tuning":true},"description":"DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.","release_date":"2025-11-20","parameters":671,"page_url":"https://ollama.com/library/deepseek-v3.1","tagline":"DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.","context_window":163840,"license":"open-weight","open_weight":true,"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-v3.2","name":"deepseek-v3.2","created_by":"deepseek","source":"official","last_updated":"2026-05-29","family":"deepseek","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"description":"DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance.","release_date":"2025-12-02","page_url":"https://ollama.com/library/deepseek-v3.2","tagline":"DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance.","context_window":163840,"license":"open-weight","open_weight":true,"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"deepseek-v4-flash","name":"deepseek-v4-flash","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Flash is a preview of the DeepSeek-V4 series, a Mixture-of-Experts model with 284B total parameters and 13B activated, built for efficient reasoning across a 1M-token context window.","release_date":"2026-04-24","reasoning_tokens":true,"license":"open-weight","parameters":284,"page_url":"https://ollama.com/library/deepseek-v4-flash","open_weight":true,"model_type":"chat","tagline":"DeepSeek-V4-Flash is a preview of the DeepSeek-V4 series, a Mixture-of-Experts model with 284B total parameters and 13B activated, built for efficient reasoning across a 1M-token context window.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","context_window":1000000,"max_output_tokens":384000},{"id":"deepseek-v4-pro","name":"deepseek-v4-pro","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a large context window and three reasoning modes.","release_date":"2026-04-24","reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/deepseek-v4-pro","open_weight":true,"model_type":"chat","tagline":"DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a 1M-token context window and three reasoning modes.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18","context_window":1000000,"max_output_tokens":384000},{"id":"devstral-2","name":"devstral-2","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"code","description":"123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","release_date":"2025-12-08","parameters":123,"page_url":"https://ollama.com/library/devstral-2","tagline":"123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","context_window":262144,"license":"mit","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"devstral-small-2","name":"devstral-small-2","created_by":"mistral","source":"official","last_updated":"2026-05-06","family":"devstral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"code","description":"24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","release_date":"2025-12-09","parameters":24,"page_url":"https://ollama.com/library/devstral-small-2","tagline":"24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","context_window":393216,"license":"apache-2.0","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"gemini-3-flash-preview","name":"gemini-3-flash-preview","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemini-3","capabilities":{"streaming":true,"fine_tuning":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","description":"Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.","release_date":"2025-12-17","page_url":"https://ollama.com/library/gemini-3-flash-preview","tagline":"Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.","license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"],"context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"gemma3","name":"gemma3","created_by":"google","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","description":"The current, most capable model that runs on a single GPU.","release_date":"2025-03-12","page_url":"https://ollama.com/library/gemma3","tagline":"The current, most capable model that runs on a single GPU.","context_window":131072,"license":"gemma","parameters":4,"open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"gemma4","name":"gemma4","created_by":"google","source":"official","description":"Gemma 4 models are designed to deliver frontier-level performance at each size. They are well-suited for reasoning, agentic workflows, coding, and multimodal understanding.","context_window":131072,"license":"open-weight","page_url":"https://ollama.com/library/gemma4","open_weight":true,"model_type":"chat","tagline":"Gemma 4 models are designed to deliver frontier-level performance at each size.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true,"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-07","parameters":31,"release_date":"2026-04-02","reasoning_tokens":true},{"id":"glm-4.6","name":"glm-4.6","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"family":"glm-4.6","model_type":"chat","description":"Advanced agentic, reasoning and coding capabilities.","release_date":"2025-09-29","page_url":"https://ollama.com/library/glm-4.6","tagline":"Advanced agentic, reasoning and coding capabilities.","context_window":202752,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"glm-4.7-flash","name":"glm-4.7-flash","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"family":"glm-4.7","model_type":"chat","description":"As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.","page_url":"https://ollama.com/library/glm-4.7-flash","tagline":"As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.","context_window":202752,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"glm-4.7","name":"glm-4.7","created_by":"zhipu","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"family":"glm-4.7","model_type":"chat","description":"Advancing the Coding Capability","release_date":"2025-12-22","page_url":"https://ollama.com/library/glm-4.7","tagline":"Advancing the Coding Capability","context_window":202752,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"glm-5.1","name":"glm-5.1","created_by":"zhipu","source":"official","family":"glm-5.1","description":"GLM-5.1 is our next-generation flagship model for agentic engineering, with significantly stronger coding capabilities than its predecessor. It achieves state-of-the-art performance on SWE-Bench Pro and leads GLM-5 by a wide margin.","release_date":"2026-04-07","context_window":202752,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/glm-5.1","open_weight":true,"model_type":"chat","tagline":"GLM-5.1 is our next-generation flagship model for agentic engineering, with significantly stronger coding capabilities than its predecessor.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-08"},{"id":"glm-5.2","name":"glm-5.2","created_by":"zhipu","source":"official","family":"glm-5.2","description":"GLM-5.2 is Z.ai’s flagship model for the era of long-horizon tasks.","release_date":"2026-06-16","context_window":999424,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/glm-5.2","open_weight":true,"model_type":"chat","tagline":"GLM-5.2 is Z.ai’s flagship model for the era of long-horizon tasks.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"glm-5","name":"glm-5","created_by":"zhipu","source":"official","last_updated":"2026-04-01","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"model_type":"chat","family":"glm-5","description":"A strong reasoning and agentic model from Z.ai with 744B total parameters (40B active), built for complex systems engineering and long-horizon tasks.","release_date":"2026-02-11","page_url":"https://ollama.com/library/glm-5","tagline":"A strong reasoning and agentic model from Z.ai with 744B total parameters (40B active), built for complex systems engineering and long-horizon tasks.","context_window":202752,"parameters":744,"license":"open-weight","active_parameters":40,"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"glm-ocr","name":"glm-ocr","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","family":"glm","description":"GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.","page_url":"https://ollama.com/library/glm-ocr","tagline":"GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.","context_window":131072,"parameters":0.9,"license":"open-weight","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text","image"],"output":["text"]}},{"id":"gpt-oss","name":"gpt-oss","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-oss","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","description":"OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.","release_date":"2025-08-05","page_url":"https://ollama.com/library/gpt-oss","tagline":"OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.","context_window":131072,"license":"apache-2.0","parameters":20,"open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"granite4.1","name":"granite4.1","created_by":"ibm","source":"official","context_window":131072,"license":"apache-2.0","page_url":"https://ollama.com/library/granite4.1","open_weight":true,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-10","description":"IBM Granite Models are a family of enterprise-ready, open foundation models that support multilingual capabilities, coding, retrieval-augmented generation (RAG), tool use, and structured JSON output. Released under Apache 2.0 license.","tagline":"IBM Granite Models are a family of enterprise-ready, open foundation models that support multilingual capabilities, coding, retrieval-augmented generation (RAG), tool use, and structured JSON output."},{"id":"granite4","name":"granite4","created_by":"ibm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.","page_url":"https://ollama.com/library/granite4","tagline":"Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.","context_window":131072,"license":"apache-2.0","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"kimi-k2-thinking","name":"kimi-k2-thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"model_type":"chat","family":"kimi-k2","description":"Kimi K2 Thinking, Moonshot AI's best open-source thinking model.","release_date":"2025-11-06","page_url":"https://ollama.com/library/kimi-k2-thinking","tagline":"Kimi K2 Thinking, Moonshot AI's best open-source thinking model.","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":32768},{"id":"kimi-k2.5","name":"kimi-k2.5","created_by":"moonshot","source":"official","last_updated":"2026-05-01","capabilities":{"streaming":true,"fine_tuning":true,"vision":true,"tool_call":true,"reasoning":true},"family":"kimi-k2.5","model_type":"chat","description":"Kimi K2.5 is an open-source, native multimodal agentic model that seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic paradigms.","release_date":"2026-01-26","page_url":"https://ollama.com/library/kimi-k2.5","tagline":"Kimi K2.5 is an open-source, native multimodal agentic model that seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as con","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"],"max_output_tokens":32768},{"id":"kimi-k2.6","name":"kimi-k2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is an open-source, native multimodal agentic model that advances practical capabilities in long-horizon coding, coding-driven design, proactive autonomous execution, and swarm-based task orchestration.","release_date":"2026-03-31","context_window":262144,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/kimi-k2.6","open_weight":true,"model_type":"chat","tagline":"Kimi K2.6 is an open-source, native multimodal agentic model that advances practical capabilities in long-horizon coding, coding-driven design, proactive autonomous execution, and swarm-based task orc","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"kimi-k2.7-code","name":"kimi-k2.7-code","created_by":"moonshot","source":"official","family":"kimi-k2.7","description":"Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built upon Kimi K2.6, with substantial improvements on real-world long-horizon coding tasks and roughly 30% lower thinking-token usage.","release_date":"2026-06-12","context_window":262144,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/kimi-k2.7-code","open_weight":true,"model_type":"chat","tagline":"Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built upon Kimi K2.6, with substantial improvements on real-world long-horizon coding tasks and roughly 30% lower thinking-token usage.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"kimi-k2","name":"kimi-k2","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","family":"kimi-k2","description":"A state-of-the-art mixture-of-experts (MoE) language model. Kimi K2-Instruct-0905 demonstrates significant improvements in performance on public benchmarks and real-world coding agent tasks.","release_date":"2025-09-05","parameters":1000,"page_url":"https://ollama.com/library/kimi-k2","tagline":"A state-of-the-art mixture-of-experts (MoE) language model.","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"laguna-xs.2","name":"laguna-xs.2","created_by":"unknown","source":"official","description":"Laguna XS.2 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.","context_window":131072,"reasoning_tokens":true,"license":"apache-2.0","parameters":33,"page_url":"https://ollama.com/library/laguna-xs.2","open_weight":true,"tagline":"Laguna XS.2 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"lfm2.5-thinking","name":"lfm2.5-thinking","created_by":"liquid","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"LFM2.5 is a new family of hybrid models designed for on-device deployment.","page_url":"https://ollama.com/library/lfm2.5-thinking","tagline":"LFM2.5 is a new family of hybrid models designed for on-device deployment.","context_window":128000,"parameters":1.2,"license":"open-weight","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"lfm2.5","name":"lfm2.5","created_by":"liquid","source":"official","description":"LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.","context_window":128000,"reasoning_tokens":true,"license":"open-weight","parameters":8,"page_url":"https://ollama.com/library/lfm2.5","open_weight":true,"model_type":"chat","tagline":"LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"lfm2","name":"lfm2","created_by":"liquid","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"LFM2 is a family of hybrid models designed for on-device deployment. LFM2-24B-A2B is the largest model in the family, scaling the architecture to 24 billion parameters while keeping inference efficient.","page_url":"https://ollama.com/library/lfm2","tagline":"LFM2 is a family of hybrid models designed for on-device deployment.","context_window":32768,"parameters":24,"license":"open-weight","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"medgemma","name":"medgemma","created_by":"google","source":"official","description":"MedGemma is a collection of Gemma 3 variants that are trained for performance on medical text and image comprehension.","context_window":131072,"license":"open-weight","parameters":4,"page_url":"https://ollama.com/library/medgemma","open_weight":true,"tagline":"MedGemma is a collection of Gemma 3 variants that are trained for performance on medical text and image comprehension.","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-04-16"},{"id":"medgemma1.5","name":"medgemma1.5","created_by":"google","source":"official","description":"MedGemma 1.5 4B is an updated version of the MedGemma 4B model.","context_window":131072,"license":"open-weight","parameters":4,"page_url":"https://ollama.com/library/medgemma1.5","open_weight":true,"tagline":"MedGemma 1.5 4B is an updated version of the MedGemma 4B model.","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-04-16"},{"id":"minicpm-v4.5","name":"minicpm-v4.5","created_by":"unknown","source":"official","description":"A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone","context_window":40960,"license":"open-weight","parameters":8,"page_url":"https://ollama.com/library/minicpm-v4.5","open_weight":true,"tagline":"A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"minicpm-v4.6","name":"minicpm-v4.6","created_by":"unknown","source":"official","context_window":262144,"license":"open-weight","parameters":1,"page_url":"https://ollama.com/library/minicpm-v4.6","open_weight":true,"capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-18","description":"A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone","tagline":"A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone"},{"id":"minimax-m2.1","name":"minimax-m2.1","created_by":"minimax","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true},"family":"minimax","model_type":"chat","description":"Exceptional multilingual capabilities to elevate code engineering","release_date":"2025-12-20","page_url":"https://ollama.com/library/minimax-m2.1","tagline":"Exceptional multilingual capabilities to elevate code engineering","context_window":204800,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"minimax-m2.5","name":"minimax-m2.5","created_by":"minimax","source":"official","last_updated":"2026-04-01","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"family":"minimax","model_type":"chat","description":"MiniMax-M2.5 is a state-of-the-art large language model designed for real-world productivity and coding tasks.","release_date":"2026-02-12","page_url":"https://ollama.com/library/minimax-m2.5","tagline":"MiniMax-M2.5 is a state-of-the-art large language model designed for real-world productivity and coding tasks.","context_window":202752,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"minimax-m2.7","name":"minimax-m2.7","created_by":"minimax","source":"official","last_updated":"2026-04-01","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"family":"minimax","model_type":"chat","description":"MiniMax's M2-series model for coding, agentic workflows, and professional productivity.","release_date":"2026-03-18","page_url":"https://ollama.com/library/minimax-m2.7","tagline":"MiniMax's M2-series model for coding, agentic workflows, and professional productivity.","context_window":204800,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"minimax-m2","name":"minimax-m2","created_by":"minimax","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true},"model_type":"chat","family":"minimax","description":"MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.","release_date":"2025-10-27","page_url":"https://ollama.com/library/minimax-m2","tagline":"MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.","context_window":204800,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"minimax-m3","name":"minimax-m3","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M3: Coding & Agentic Frontier. 1M context window. Native Multimodality.","release_date":"2026-06-01","context_window":524288,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/minimax-m3","open_weight":true,"model_type":"chat","tagline":"MiniMax M3: Coding & Agentic Frontier.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"ministral-3","name":"ministral-3","created_by":"mistral","source":"official","last_updated":"2026-05-07","family":"ministral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","description":"The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.","release_date":"2025-12-02","page_url":"https://ollama.com/library/ministral-3","tagline":"The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.","context_window":262144,"license":"apache-2.0","parameters":8,"open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral-large-3","name":"mistral-large-3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-large","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","description":"A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.","release_date":"2025-12-02","parameters":675,"page_url":"https://ollama.com/library/mistral-large-3","tagline":"A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.","context_window":262144,"license":"apache-2.0","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral-medium-3.5","name":"mistral-medium-3.5","created_by":"mistral","source":"official","family":"mistral-medium","description":"Mistral Medium 3.5 is the first flagship model of Mistral AI that merged instruction-following, reasoning, and coding in a single set of 128B weights.","context_window":262144,"reasoning_tokens":true,"license":"mit","parameters":128,"page_url":"https://ollama.com/library/mistral-medium-3.5","open_weight":true,"model_type":"chat","tagline":"Mistral Medium 3.5 is the first flagship model of Mistral AI that merged instruction-following, reasoning, and coding in a single set of 128B weights.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"nemotron-3-nano","name":"nemotron-3-nano","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"Nemotron-3-Nano is a new Standard for Efficient, Open, and Intelligent Agentic Models, now updated with a 4B parameter count model.","release_date":"2025-12-15","parameters":30,"page_url":"https://ollama.com/library/nemotron-3-nano","tagline":"Nemotron-3-Nano is a new Standard for Efficient, Open, and Intelligent Agentic Models, now updated with a 4B parameter count model.","context_window":262144,"license":"nvidia-open","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron-3-super","name":"nemotron-3-super","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"NVIDIA Nemotron 3 Super is a 120B open MoE model activating just 12B parameters to deliver maximum compute efficiency and accuracy for complex multi-agent applications.","release_date":"2026-03-11","page_url":"https://ollama.com/library/nemotron-3-super","tagline":"NVIDIA Nemotron 3 Super is a 120B open MoE model activating just 12B parameters to deliver maximum compute efficiency and accuracy for complex multi-agent applications.","context_window":262144,"parameters":120,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron-3-ultra","name":"nemotron-3-ultra","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.","release_date":"2026-06-04","context_window":262144,"reasoning_tokens":true,"license":"open-weight","parameters":550,"active_parameters":55,"page_url":"https://ollama.com/library/nemotron-3-ultra","open_weight":true,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nemotron-cascade-2","name":"nemotron-cascade-2","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.","page_url":"https://ollama.com/library/nemotron-cascade-2","tagline":"An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.","context_window":262144,"parameters":30,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron3","name":"nemotron3","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Nano Omni is a multimodal large language model that unifies video, audio, image, and text understanding to support enterprise-grade Q&A, summarization, transcription, and document intelligence workflows.","context_window":131072,"reasoning_tokens":true,"license":"open-weight","parameters":33,"page_url":"https://ollama.com/library/nemotron3","open_weight":true,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Nano Omni is a multimodal large language model that unifies video, audio, image, and text understanding to support enterprise-grade Q&A, summarization, transcription, and documen","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"north-mini-code-1.0","name":"north-mini-code-1.0","created_by":"unknown","source":"official","description":"North Mini Code is Cohere's first model for developers — a 30B Mixture-of-Experts model with 3B active parameters, built for agentic software engineering.","context_window":499712,"reasoning_tokens":true,"license":"apache-2.0","page_url":"https://ollama.com/library/north-mini-code-1.0","open_weight":true,"tagline":"North Mini Code is Cohere's first model for developers — a 30B Mixture-of-Experts model with 3B active parameters, built for agentic software engineering.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"olmo-3","name":"olmo-3","created_by":"unknown","source":"official","description":"Olmo is a series of Open language models designed to enable the science of language models. These models are pre-trained on the Dolma 3 dataset and post-trained on the Dolci datasets.","context_window":65536,"license":"open-weight","parameters":7,"page_url":"https://ollama.com/library/olmo-3","open_weight":true,"tagline":"Olmo is a series of Open language models designed to enable the science of language models.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-01"},{"id":"qwen3-coder-next","name":"qwen3-coder-next","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","description":"Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.","release_date":"2025-02-04","page_url":"https://ollama.com/library/qwen3-coder-next","tagline":"Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.","context_window":262144,"license":"open-weight","parameters":80,"active_parameters":3,"open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536},{"id":"qwen3-coder","name":"qwen3-coder","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"code","description":"Alibaba's performant long context models for agentic and coding tasks.","release_date":"2025-07-22","parameters":480,"page_url":"https://ollama.com/library/qwen3-coder","tagline":"Alibaba's performant long context models for agentic and coding tasks.","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen3-embedding","name":"qwen3-embedding","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true},"model_type":"embed"},{"id":"qwen3-next","name":"qwen3-next","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.","release_date":"2025-09-10","parameters":80,"page_url":"https://ollama.com/library/qwen3-next","tagline":"The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.","context_window":262144,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen3-vl","name":"qwen3-vl","created_by":"qwen","source":"official","last_updated":"2026-04-12","family":"qwen3","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"The most powerful vision-language model in the Qwen model family to date.","release_date":"2025-09-22","parameters":235,"page_url":"https://ollama.com/library/qwen3-vl","tagline":"The most powerful vision-language model in the Qwen model family to date.","context_window":262144,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen3.5","name":"qwen3.5","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.","release_date":"2026-02-16","parameters":397,"page_url":"https://ollama.com/library/qwen3.5","tagline":"Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.","context_window":262144,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text","image"],"output":["text"]}},{"id":"qwen3.6","name":"qwen3.6","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6 delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.","context_window":262144,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/qwen3.6","open_weight":true,"model_type":"chat","tagline":"Qwen3.6 delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-21","parameters":35},{"id":"rnj-1","name":"rnj-1","created_by":"essentialai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models.","release_date":"2025-12-09","parameters":8,"page_url":"https://ollama.com/library/rnj-1","tagline":"Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models.","context_window":32768,"license":"open-weight","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"translategemma","name":"translategemma","created_by":"google","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"model_type":"translation","description":"A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.","page_url":"https://ollama.com/library/translategemma","tagline":"A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.","context_window":131072,"parameters":4,"license":"open-weight","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]}}]};
+export const provider: ProviderWithModels = {"id":"ollama","name":"Ollama","region":"US","headquarters":"Palo Alto, CA","founded":2023,"url":"https://ollama.com","api_url":"http://localhost:11434/api","docs_url":"https://github.com/ollama/ollama/blob/main/docs/api.md","pricing_url":"https://ollama.com","description":"Run open-source LLMs locally on your machine with a simple CLI.","type":"aggregator","openai_compatible":true,"free_tier":true,"github_url":"https://github.com/ollama/ollama","models_url":"https://ollama.com/library","twitter_url":"https://x.com/ollama","discord_url":"https://discord.com/invite/ollama","blog_url":"https://ollama.com/blog","support_url":"https://github.com/ollama/ollama","icon":"","models":[{"id":"cogito-2.1","name":"cogito-2.1","created_by":"deepcogito","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"description":"The Cogito v2.1 LLMs are instruction tuned generative models. All models are released under MIT license for commercial use.","release_date":"2025-11-19","parameters":671,"page_url":"https://ollama.com/library/cogito-2.1","tagline":"The Cogito v2.1 LLMs are instruction tuned generative models.","context_window":163840,"license":"mit","open_weight":true,"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-v3.1","name":"deepseek-v3.1","created_by":"deepseek","source":"official","last_updated":"2026-03-24","family":"deepseek","capabilities":{"streaming":true,"fine_tuning":true},"description":"DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.","release_date":"2025-11-20","parameters":671,"page_url":"https://ollama.com/library/deepseek-v3.1","tagline":"DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.","context_window":163840,"license":"open-weight","open_weight":true,"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek-v3.2","name":"deepseek-v3.2","created_by":"deepseek","source":"official","last_updated":"2026-05-29","family":"deepseek","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"description":"DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance.","release_date":"2025-12-02","page_url":"https://ollama.com/library/deepseek-v3.2","tagline":"DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance.","context_window":163840,"license":"open-weight","open_weight":true,"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"deepseek-v4-flash","name":"deepseek-v4-flash","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Flash is a preview of the DeepSeek-V4 series, a Mixture-of-Experts model with 284B total parameters and 13B activated, built for efficient reasoning across a 1M-token context window.","release_date":"2026-04-24","reasoning_tokens":true,"license":"open-weight","parameters":284,"page_url":"https://ollama.com/library/deepseek-v4-flash","open_weight":true,"model_type":"chat","tagline":"DeepSeek-V4-Flash is a preview of the DeepSeek-V4 series, a Mixture-of-Experts model with 284B total parameters and 13B activated, built for efficient reasoning across a 1M-token context window.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","context_window":1000000,"max_output_tokens":384000},{"id":"deepseek-v4-pro","name":"deepseek-v4-pro","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a large context window and three reasoning modes.","release_date":"2026-04-24","reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/deepseek-v4-pro","open_weight":true,"model_type":"chat","tagline":"DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a 1M-token context window and three reasoning modes.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18","context_window":1000000,"max_output_tokens":384000},{"id":"devstral-2","name":"devstral-2","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"code","description":"123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","release_date":"2025-12-08","parameters":123,"page_url":"https://ollama.com/library/devstral-2","tagline":"123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","context_window":262144,"license":"mit","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"devstral-small-2","name":"devstral-small-2","created_by":"mistral","source":"official","last_updated":"2026-05-06","family":"devstral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"code","description":"24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","release_date":"2025-12-09","parameters":24,"page_url":"https://ollama.com/library/devstral-small-2","tagline":"24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","context_window":393216,"license":"apache-2.0","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"gemini-3-flash-preview","name":"gemini-3-flash-preview","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemini-3","capabilities":{"streaming":true,"fine_tuning":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"chat","description":"Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.","release_date":"2025-12-17","page_url":"https://ollama.com/library/gemini-3-flash-preview","tagline":"Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.","license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"],"context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"gemma3","name":"gemma3","created_by":"google","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","description":"The current, most capable model that runs on a single GPU.","release_date":"2025-03-12","page_url":"https://ollama.com/library/gemma3","tagline":"The current, most capable model that runs on a single GPU.","context_window":131072,"license":"gemma","parameters":4,"open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"gemma4","name":"gemma4","created_by":"google","source":"official","description":"Gemma 4 models are designed to deliver frontier-level performance at each size. They are well-suited for reasoning, agentic workflows, coding, and multimodal understanding.","context_window":131072,"license":"open-weight","page_url":"https://ollama.com/library/gemma4","open_weight":true,"model_type":"chat","tagline":"Gemma 4 models are designed to deliver frontier-level performance at each size.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true,"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-07","parameters":31,"release_date":"2026-04-02","reasoning_tokens":true},{"id":"glm-4.6","name":"glm-4.6","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"family":"glm-4.6","model_type":"chat","description":"Advanced agentic, reasoning and coding capabilities.","release_date":"2025-09-29","page_url":"https://ollama.com/library/glm-4.6","tagline":"Advanced agentic, reasoning and coding capabilities.","context_window":202752,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"glm-4.7-flash","name":"glm-4.7-flash","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"family":"glm-4.7","model_type":"chat","description":"As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.","page_url":"https://ollama.com/library/glm-4.7-flash","tagline":"As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.","context_window":202752,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"glm-4.7","name":"glm-4.7","created_by":"zhipu","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"family":"glm-4.7","model_type":"chat","description":"Advancing the Coding Capability","release_date":"2025-12-22","page_url":"https://ollama.com/library/glm-4.7","tagline":"Advancing the Coding Capability","context_window":202752,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"glm-5.1","name":"glm-5.1","created_by":"zhipu","source":"official","family":"glm-5.1","description":"GLM-5.1 is our next-generation flagship model for agentic engineering, with significantly stronger coding capabilities than its predecessor. It achieves state-of-the-art performance on SWE-Bench Pro and leads GLM-5 by a wide margin.","release_date":"2026-04-07","context_window":202752,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/glm-5.1","open_weight":true,"model_type":"chat","tagline":"GLM-5.1 is our next-generation flagship model for agentic engineering, with significantly stronger coding capabilities than its predecessor.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-08"},{"id":"glm-5.2","name":"glm-5.2","created_by":"zhipu","source":"official","family":"glm-5.2","description":"GLM-5.2 is Z.ai’s flagship model for the era of long-horizon tasks.","release_date":"2026-06-16","context_window":999424,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/glm-5.2","open_weight":true,"model_type":"chat","tagline":"GLM-5.2 is Z.ai’s flagship model for the era of long-horizon tasks.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"glm-5","name":"glm-5","created_by":"zhipu","source":"official","last_updated":"2026-04-01","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"model_type":"chat","family":"glm-5","description":"A strong reasoning and agentic model from Z.ai with 744B total parameters (40B active), built for complex systems engineering and long-horizon tasks.","release_date":"2026-02-11","page_url":"https://ollama.com/library/glm-5","tagline":"A strong reasoning and agentic model from Z.ai with 744B total parameters (40B active), built for complex systems engineering and long-horizon tasks.","context_window":202752,"parameters":744,"license":"open-weight","active_parameters":40,"open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"glm-ocr","name":"glm-ocr","created_by":"zhipu","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","family":"glm","description":"GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.","page_url":"https://ollama.com/library/glm-ocr","tagline":"GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.","context_window":131072,"parameters":0.9,"license":"open-weight","open_weight":true,"tools":["function_calling"],"modalities":{"input":["text","image"],"output":["text"]}},{"id":"gpt-oss","name":"gpt-oss","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-oss","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","description":"OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.","release_date":"2025-08-05","page_url":"https://ollama.com/library/gpt-oss","tagline":"OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.","context_window":131072,"license":"apache-2.0","parameters":20,"open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"granite4.1","name":"granite4.1","created_by":"ibm","source":"official","context_window":131072,"license":"apache-2.0","page_url":"https://ollama.com/library/granite4.1","open_weight":true,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-10","description":"IBM Granite Models are a family of enterprise-ready, open foundation models that support multilingual capabilities, coding, retrieval-augmented generation (RAG), tool use, and structured JSON output. Released under Apache 2.0 license.","tagline":"IBM Granite Models are a family of enterprise-ready, open foundation models that support multilingual capabilities, coding, retrieval-augmented generation (RAG), tool use, and structured JSON output."},{"id":"granite4","name":"granite4","created_by":"ibm","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.","page_url":"https://ollama.com/library/granite4","tagline":"Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.","context_window":131072,"license":"apache-2.0","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"kimi-k2-thinking","name":"kimi-k2-thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"model_type":"chat","family":"kimi-k2","description":"Kimi K2 Thinking, Moonshot AI's best open-source thinking model.","release_date":"2025-11-06","page_url":"https://ollama.com/library/kimi-k2-thinking","tagline":"Kimi K2 Thinking, Moonshot AI's best open-source thinking model.","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":32768},{"id":"kimi-k2.5","name":"kimi-k2.5","created_by":"moonshot","source":"official","last_updated":"2026-05-01","capabilities":{"streaming":true,"fine_tuning":true,"vision":true,"tool_call":true,"reasoning":true},"family":"kimi-k2.5","model_type":"chat","description":"Kimi K2.5 is an open-source, native multimodal agentic model that seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic paradigms.","release_date":"2026-01-26","page_url":"https://ollama.com/library/kimi-k2.5","tagline":"Kimi K2.5 is an open-source, native multimodal agentic model that seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as con","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"],"max_output_tokens":32768},{"id":"kimi-k2.6","name":"kimi-k2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is an open-source, native multimodal agentic model that advances practical capabilities in long-horizon coding, coding-driven design, proactive autonomous execution, and swarm-based task orchestration.","release_date":"2026-03-31","context_window":262144,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/kimi-k2.6","open_weight":true,"model_type":"chat","tagline":"Kimi K2.6 is an open-source, native multimodal agentic model that advances practical capabilities in long-horizon coding, coding-driven design, proactive autonomous execution, and swarm-based task orc","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"kimi-k2.7-code","name":"kimi-k2.7-code","created_by":"moonshot","source":"official","family":"kimi-k2.7","description":"Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built upon Kimi K2.6, with substantial improvements on real-world long-horizon coding tasks and roughly 30% lower thinking-token usage.","release_date":"2026-06-12","context_window":262144,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/kimi-k2.7-code","open_weight":true,"model_type":"chat","tagline":"Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built upon Kimi K2.6, with substantial improvements on real-world long-horizon coding tasks and roughly 30% lower thinking-token usage.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"kimi-k2","name":"kimi-k2","created_by":"moonshot","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","family":"kimi-k2","description":"A state-of-the-art mixture-of-experts (MoE) language model. Kimi K2-Instruct-0905 demonstrates significant improvements in performance on public benchmarks and real-world coding agent tasks.","release_date":"2025-09-05","parameters":1000,"page_url":"https://ollama.com/library/kimi-k2","tagline":"A state-of-the-art mixture-of-experts (MoE) language model.","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"laguna-xs.2","name":"laguna-xs.2","created_by":"unknown","source":"official","description":"Laguna XS.2 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.","context_window":131072,"reasoning_tokens":true,"license":"apache-2.0","parameters":33,"page_url":"https://ollama.com/library/laguna-xs.2","open_weight":true,"tagline":"Laguna XS.2 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"lfm2.5-thinking","name":"lfm2.5-thinking","created_by":"liquid","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"LFM2.5 is a new family of hybrid models designed for on-device deployment.","page_url":"https://ollama.com/library/lfm2.5-thinking","tagline":"LFM2.5 is a new family of hybrid models designed for on-device deployment.","context_window":128000,"parameters":1.2,"license":"open-weight","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"lfm2.5","name":"lfm2.5","created_by":"liquid","source":"official","description":"LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.","context_window":128000,"reasoning_tokens":true,"license":"open-weight","parameters":8,"page_url":"https://ollama.com/library/lfm2.5","open_weight":true,"model_type":"chat","tagline":"LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"lfm2","name":"lfm2","created_by":"liquid","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"LFM2 is a family of hybrid models designed for on-device deployment. LFM2-24B-A2B is the largest model in the family, scaling the architecture to 24 billion parameters while keeping inference efficient.","page_url":"https://ollama.com/library/lfm2","tagline":"LFM2 is a family of hybrid models designed for on-device deployment.","context_window":32768,"parameters":24,"license":"open-weight","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"medgemma","name":"medgemma","created_by":"google","source":"official","description":"MedGemma is a collection of Gemma 3 variants that are trained for performance on medical text and image comprehension.","context_window":131072,"license":"open-weight","parameters":4,"page_url":"https://ollama.com/library/medgemma","open_weight":true,"tagline":"MedGemma is a collection of Gemma 3 variants that are trained for performance on medical text and image comprehension.","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-04-16"},{"id":"medgemma1.5","name":"medgemma1.5","created_by":"google","source":"official","description":"MedGemma 1.5 4B is an updated version of the MedGemma 4B model.","context_window":131072,"license":"open-weight","parameters":4,"page_url":"https://ollama.com/library/medgemma1.5","open_weight":true,"tagline":"MedGemma 1.5 4B is an updated version of the MedGemma 4B model.","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-04-16"},{"id":"minicpm-v4.5","name":"minicpm-v4.5","created_by":"unknown","source":"official","description":"A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone","context_window":40960,"license":"open-weight","parameters":8,"page_url":"https://ollama.com/library/minicpm-v4.5","open_weight":true,"tagline":"A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"minicpm-v4.6","name":"minicpm-v4.6","created_by":"unknown","source":"official","context_window":262144,"license":"open-weight","parameters":1,"page_url":"https://ollama.com/library/minicpm-v4.6","open_weight":true,"capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-18","description":"A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone","tagline":"A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone"},{"id":"minimax-m2.1","name":"minimax-m2.1","created_by":"minimax","source":"official","last_updated":"2026-05-23","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true},"family":"minimax","model_type":"chat","description":"Exceptional multilingual capabilities to elevate code engineering","release_date":"2025-12-20","page_url":"https://ollama.com/library/minimax-m2.1","tagline":"Exceptional multilingual capabilities to elevate code engineering","context_window":204800,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"minimax-m2.5","name":"minimax-m2.5","created_by":"minimax","source":"official","last_updated":"2026-04-01","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"family":"minimax","model_type":"chat","description":"MiniMax-M2.5 is a state-of-the-art large language model designed for real-world productivity and coding tasks.","release_date":"2026-02-12","page_url":"https://ollama.com/library/minimax-m2.5","tagline":"MiniMax-M2.5 is a state-of-the-art large language model designed for real-world productivity and coding tasks.","context_window":202752,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"minimax-m2.7","name":"minimax-m2.7","created_by":"minimax","source":"official","last_updated":"2026-04-01","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"reasoning":true},"family":"minimax","model_type":"chat","description":"MiniMax's M2-series model for coding, agentic workflows, and professional productivity.","release_date":"2026-03-18","page_url":"https://ollama.com/library/minimax-m2.7","tagline":"MiniMax's M2-series model for coding, agentic workflows, and professional productivity.","context_window":204800,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"minimax-m2","name":"minimax-m2","created_by":"minimax","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true},"model_type":"chat","family":"minimax","description":"MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.","release_date":"2025-10-27","page_url":"https://ollama.com/library/minimax-m2","tagline":"MiniMax M2 is a high-efficiency large language model built for coding and agentic workflows.","context_window":204800,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"minimax-m3","name":"minimax-m3","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax M3: Coding & Agentic Frontier. 1M context window. Native Multimodality.","release_date":"2026-06-01","context_window":524288,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/minimax-m3","open_weight":true,"model_type":"chat","tagline":"MiniMax M3: Coding & Agentic Frontier.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"ministral-3","name":"ministral-3","created_by":"mistral","source":"official","last_updated":"2026-05-07","family":"ministral","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","description":"The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.","release_date":"2025-12-02","page_url":"https://ollama.com/library/ministral-3","tagline":"The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.","context_window":262144,"license":"apache-2.0","parameters":8,"open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral-large-3","name":"mistral-large-3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-large","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"chat","description":"A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.","release_date":"2025-12-02","parameters":675,"page_url":"https://ollama.com/library/mistral-large-3","tagline":"A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.","context_window":262144,"license":"apache-2.0","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral-medium-3.5","name":"mistral-medium-3.5","created_by":"mistral","source":"official","family":"mistral-medium","description":"Mistral Medium 3.5 is the first flagship model of Mistral AI that merged instruction-following, reasoning, and coding in a single set of 128B weights.","context_window":262144,"reasoning_tokens":true,"license":"mit","parameters":128,"page_url":"https://ollama.com/library/mistral-medium-3.5","open_weight":true,"model_type":"chat","tagline":"Mistral Medium 3.5 is the first flagship model of Mistral AI that merged instruction-following, reasoning, and coding in a single set of 128B weights.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"nemotron-3-nano","name":"nemotron-3-nano","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"Nemotron-3-Nano is a new Standard for Efficient, Open, and Intelligent Agentic Models, now updated with a 4B parameter count model.","release_date":"2025-12-15","parameters":30,"page_url":"https://ollama.com/library/nemotron-3-nano","tagline":"Nemotron-3-Nano is a new Standard for Efficient, Open, and Intelligent Agentic Models, now updated with a 4B parameter count model.","context_window":262144,"license":"nvidia-open","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron-3-super","name":"nemotron-3-super","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"NVIDIA Nemotron 3 Super is a 120B open MoE model activating just 12B parameters to deliver maximum compute efficiency and accuracy for complex multi-agent applications.","release_date":"2026-03-11","page_url":"https://ollama.com/library/nemotron-3-super","tagline":"NVIDIA Nemotron 3 Super is a 120B open MoE model activating just 12B parameters to deliver maximum compute efficiency and accuracy for complex multi-agent applications.","context_window":262144,"parameters":120,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron-3-ultra","name":"nemotron-3-ultra","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.","release_date":"2026-06-04","context_window":262144,"reasoning_tokens":true,"license":"open-weight","parameters":550,"active_parameters":55,"page_url":"https://ollama.com/library/nemotron-3-ultra","open_weight":true,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nemotron-cascade-2","name":"nemotron-cascade-2","created_by":"nvidia","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.","page_url":"https://ollama.com/library/nemotron-cascade-2","tagline":"An open 30B MoE model from NVIDIA with 3B activated parameters that delivers strong reasoning and agentic capabilities.","context_window":262144,"parameters":30,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron3","name":"nemotron3","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Nano Omni is a multimodal large language model that unifies video, audio, image, and text understanding to support enterprise-grade Q&A, summarization, transcription, and document intelligence workflows.","context_window":131072,"reasoning_tokens":true,"license":"open-weight","parameters":33,"page_url":"https://ollama.com/library/nemotron3","open_weight":true,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Nano Omni is a multimodal large language model that unifies video, audio, image, and text understanding to support enterprise-grade Q&A, summarization, transcription, and documen","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"north-mini-code-1.0","name":"north-mini-code-1.0","created_by":"unknown","source":"official","description":"North Mini Code is Cohere's first model for developers — a 30B Mixture-of-Experts model with 3B active parameters, built for agentic software engineering.","context_window":499712,"reasoning_tokens":true,"license":"apache-2.0","page_url":"https://ollama.com/library/north-mini-code-1.0","open_weight":true,"tagline":"North Mini Code is Cohere's first model for developers — a 30B Mixture-of-Experts model with 3B active parameters, built for agentic software engineering.","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"olmo-3","name":"olmo-3","created_by":"unknown","source":"official","description":"Olmo is a series of Open language models designed to enable the science of language models. These models are pre-trained on the Dolma 3 dataset and post-trained on the Dolci datasets.","context_window":65536,"license":"open-weight","parameters":7,"page_url":"https://ollama.com/library/olmo-3","open_weight":true,"tagline":"Olmo is a series of Open language models designed to enable the science of language models.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-01"},{"id":"ornith","name":"ornith","created_by":"unknown","source":"official","description":"A self-improving family of open-source models for agentic coding","context_window":262144,"license":"mit","parameters":9,"page_url":"https://ollama.com/library/ornith","open_weight":true,"tagline":"A self-improving family of open-source models for agentic coding","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29"},{"id":"qwen3-coder-next","name":"qwen3-coder-next","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"model_type":"code","description":"Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.","release_date":"2025-02-04","page_url":"https://ollama.com/library/qwen3-coder-next","tagline":"Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.","context_window":262144,"license":"open-weight","parameters":80,"active_parameters":3,"open_weight":true,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536},{"id":"qwen3-coder","name":"qwen3-coder","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","capabilities":{"streaming":true,"fine_tuning":true},"model_type":"code","description":"Alibaba's performant long context models for agentic and coding tasks.","release_date":"2025-07-22","parameters":480,"page_url":"https://ollama.com/library/qwen3-coder","tagline":"Alibaba's performant long context models for agentic and coding tasks.","context_window":262144,"license":"open-weight","open_weight":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen3-embedding","name":"qwen3-embedding","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true},"model_type":"embed"},{"id":"qwen3-next","name":"qwen3-next","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.","release_date":"2025-09-10","parameters":80,"page_url":"https://ollama.com/library/qwen3-next","tagline":"The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.","context_window":262144,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen3-vl","name":"qwen3-vl","created_by":"qwen","source":"official","last_updated":"2026-04-12","family":"qwen3","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"The most powerful vision-language model in the Qwen model family to date.","release_date":"2025-09-22","parameters":235,"page_url":"https://ollama.com/library/qwen3-vl","tagline":"The most powerful vision-language model in the Qwen model family to date.","context_window":262144,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text"],"output":["text"]}},{"id":"qwen3.5","name":"qwen3.5","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"model_type":"chat","description":"Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.","release_date":"2026-02-16","parameters":397,"page_url":"https://ollama.com/library/qwen3.5","tagline":"Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.","context_window":262144,"license":"open-weight","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true,"modalities":{"input":["text","image"],"output":["text"]}},{"id":"qwen3.6","name":"qwen3.6","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6 delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.","context_window":262144,"reasoning_tokens":true,"license":"open-weight","page_url":"https://ollama.com/library/qwen3.6","open_weight":true,"model_type":"chat","tagline":"Qwen3.6 delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-21","parameters":35},{"id":"rnj-1","name":"rnj-1","created_by":"essentialai","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"description":"Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models.","release_date":"2025-12-09","parameters":8,"page_url":"https://ollama.com/library/rnj-1","tagline":"Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models.","context_window":32768,"license":"open-weight","open_weight":true,"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"translategemma","name":"translategemma","created_by":"google","source":"official","last_updated":"2026-03-24","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"model_type":"translation","description":"A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.","page_url":"https://ollama.com/library/translategemma","tagline":"A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.","context_window":131072,"parameters":4,"license":"open-weight","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]}}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/opencode.ts b/packages/npm/src/providers/opencode.ts
index 7f7071a16..6c020b52b 100644
--- a/packages/npm/src/providers/opencode.ts
+++ b/packages/npm/src/providers/opencode.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"opencode","name":"OpenCode Zen","region":"US","headquarters":"San Francisco, CA","founded":2025,"url":"https://opencode.ai","api_url":"https://opencode.ai/zen/v1","docs_url":"https://opencode.ai/docs/zen","pricing_url":"https://opencode.ai/docs/zen","description":"AI coding assistant platform providing access to frontier models for software development.","type":"cloud","openai_compatible":true,"free_tier":true,"twitter_url":"https://x.com/anomalyco","discord_url":"https://discord.com/invite/opencode","terms_url":"https://opencode.ai/legal/terms-of-service","support_url":"https://github.com/opencode-ai/opencode","icon":"","models":[{"id":"big-pickle","name":"Big Pickle","created_by":"opencode","source":"official","last_updated":"2026-03-21","status":"active","capabilities":{"streaming":true}},{"id":"claude-3-5-haiku","name":"Claude Haiku 3.5","created_by":"anthropic","source":"official","last_updated":"2026-04-26","family":"claude-haiku","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true},"pricing":{"input":0.8,"output":4,"cached_input":0.08,"cache_write":1},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-02-16","license":"proprietary","open_weight":false},{"id":"claude-fable-5","name":"Claude Fable 5","created_by":"anthropic","source":"official","family":"claude-fable","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":50,"cached_input":1,"cache_write":12.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","tagline":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","context_window":1000000,"max_output_tokens":128000,"license":"proprietary","open_weight":false},{"id":"claude-haiku-4-5","name":"Claude Haiku 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":1,"output":5,"cached_input":0.1,"cache_write":1.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The fastest model with near-frontier intelligence","tagline":"The fastest model with near-frontier intelligence","context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-02","training_data_cutoff":"2025-07","license":"proprietary","open_weight":false},{"id":"claude-opus-4-1","name":"Claude Opus 4.1","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":15,"output":75,"cached_input":1.5,"cache_write":18.75},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":32000,"knowledge_cutoff":"2025-01","training_data_cutoff":"2025-03","license":"proprietary","open_weight":false},{"id":"claude-opus-4-5","name":"Claude Opus 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"claude-opus-4-6","name":"Claude Opus 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The most intelligent broadly available model for agents and coding","tagline":"The most intelligent model for building agents and coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"claude-opus-4-7","name":"Claude Opus 4.7","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-17","description":"Our most capable generally available model for complex reasoning and agentic coding","tagline":"Our most capable generally available model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"claude-opus-4-8","name":"Claude Opus 4.8","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-29","description":"Anthropic's most capable Opus-tier model for complex reasoning and agentic coding","tagline":"Anthropic's most capable model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4-5","name":"Claude Sonnet 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-01","training_data_cutoff":"2025-07","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.3,"cache_write":3.75},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The best combination of speed and intelligence","tagline":"The best combination of speed and intelligence","context_window":1000000,"max_output_tokens":64000,"knowledge_cutoff":"2025-08","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4","name":"Claude Sonnet 4","created_by":"anthropic","source":"official","last_updated":"2026-04-26","family":"claude-sonnet","status":"deprecated","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-06-15"},{"id":"deepseek-v4-flash-free","name":"DeepSeek V4 Flash Free","created_by":"unknown","source":"official","family":"deepseek","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-14"},{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","created_by":"unknown","source":"official","family":"deepseek","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.14,"output":0.28,"cached_input":0.028},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","created_by":"unknown","source":"official","family":"deepseek","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.74,"output":3.48,"cached_input":0.145},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"gemini-3-flash","name":"Gemini 3 Flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","status":"active","capabilities":{"streaming":true},"pricing":{"input":0.5,"output":3,"cached_input":0.05},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"gemini-3.1-pro","name":"Gemini 3.1 Pro","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","status":"active","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"gemini-3.5-flash","name":"Gemini 3.5 Flash","created_by":"google","source":"official","family":"gemini-3.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":1.5,"output":9,"cached_input":0.15},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-23","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"glm-5.1","name":"GLM 5.1","created_by":"zhipu","source":"official","family":"glm-5.1","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.4,"output":4.4,"cached_input":0.26},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-08"},{"id":"glm-5","name":"GLM 5","created_by":"zhipu","source":"official","last_updated":"2026-04-26","status":"deprecated","capabilities":{"streaming":true},"pricing":{"input":1,"output":3.2,"cached_input":0.2},"model_type":"chat","family":"glm-5","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-05-14"},{"id":"gpt-5-codex","name":"GPT 5 Codex","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5 optimized for agentic coding in Codex","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5-nano","name":"GPT 5 Nano","created_by":"openai","source":"official","last_updated":"2026-05-07","family":"gpt-5","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.4,"cached_input":0.005},"description":"GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for","tagline":"Fastest, most cost-efficient version of GPT-5","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"gpt-5.1-codex-max","name":"GPT 5.1 Codex Max","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.1","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT‑5.1-Codex-Max is purpose-built for agentic coding.","tagline":"A version of GPT-5.1-codex optimized for long running tasks.","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.1-codex-mini","name":"GPT 5.1 Codex Mini","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.1","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5.1-Codex mini is a smaller, more cost-effective, less-capable version of GPT-5.1-Codex.","tagline":"Smaller, more cost-effective, less-capable version of GPT-5.1-Codex","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.1-codex","name":"GPT 5.1 Codex","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.1","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5.1-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5.1 optimized for agentic coding in Codex.","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.1","name":"GPT 5.1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1 is our flagship model for coding and agentic tasks with configurable reasoning and non-reasoning effort.","tagline":"The best model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.2-codex","name":"GPT 5.2 Codex","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.2","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5.2-Codex is an upgraded version of GPT-5.2 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.2-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.2-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"Our most intelligent coding model optimized for long-horizon, agentic coding tasks.","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.2","name":"GPT 5.2","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.3-codex-spark","name":"GPT 5.3 Codex Spark","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","status":"active","capabilities":{"streaming":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"gpt-5.3-codex","name":"GPT 5.3 Codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.3-Codex is optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.3-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.3-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"The most capable agentic coding model to date.","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4-mini","name":"GPT 5.4 Mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5,"cached_input":0.075},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 mini brings the strengths of GPT-5.4 to a faster, more efficient","tagline":"Our strongest mini model yet for coding, computer use, and subagents","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4-nano","name":"GPT 5.4 Nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25,"cached_input":0.02},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 nano is designed for tasks where speed and cost matter most like","tagline":"Our cheapest GPT-5.4-class model for simple high-volume tasks","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4-pro","name":"GPT 5.4 Pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false},"pricing":{"input":30,"output":180,"cached_input":30},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4","name":"GPT 5.4","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15,"cached_input":0.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 is our frontier model for complex professional work.","tagline":"A more affordable model for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.5-pro","name":"GPT 5.5 Pro","created_by":"openai","source":"official","family":"gpt-5.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"reasoning":true,"batch":true},"pricing":{"input":30,"output":180,"cached_input":30},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"gpt-5.5","name":"GPT 5.5","created_by":"openai","source":"official","family":"gpt-5.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"reasoning":true,"batch":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","description":"GPT-5.5 is our newest frontier model for the most complex professional work.","tagline":"A new class of intelligence for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"gpt-5","name":"GPT 5","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"grok-build-0.1","name":"Grok Build 0.1","created_by":"unknown","source":"official","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"output":2,"cached_input":0.2,"input":1},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02"},{"id":"hy3-preview-free","name":"Hy3 Preview Free","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-04-24"},{"id":"kimi-k2.5","name":"Kimi K2.5","created_by":"moonshot","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":3,"cached_input":0.1},"family":"kimi-k2.5","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","tagline":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","context_window":256000,"max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"kimi-k2.6","name":"Kimi K2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.95,"output":4,"cached_input":0.16},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21"},{"id":"ling-2.6-flash","name":"Ling 2.6 Flash","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-04-22"},{"id":"mimo-v2-omni-free","name":"MiMo V2 Omni Free","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"model_type":"chat","family":"mimo","modalities":{"input":["text"],"output":["text"]}},{"id":"mimo-v2-pro-free","name":"MiMo V2 Pro Free","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"model_type":"chat","family":"mimo","modalities":{"input":["text"],"output":["text"]}},{"id":"mimo-v2.5-free","name":"MiMo-V2.5 Free","created_by":"xiaomi","source":"official","family":"mimo","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-29"},{"id":"minimax-m2.5-free","name":"MiniMax M2.5 Free","created_by":"minimax","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"minimax-m2.5","name":"MiniMax M2.5","created_by":"minimax","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06,"cache_write":0.375},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Peak Performance. Ultimate Value. Master the Complex","tagline":"Peak Performance.","context_window":204800,"license":"mit","open_weight":true},{"id":"minimax-m2.7","name":"MiniMax M2.7","created_by":"minimax","source":"official","family":"minimax","status":"active","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06,"cache_write":0.375},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","context_window":204800,"license":"proprietary","open_weight":false},{"id":"nemotron-3-super-free","name":"Nemotron 3 Super Free","created_by":"nvidia","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron-3-ultra-free","name":"Nemotron 3 Ultra Free","created_by":"nvidia","source":"official","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10"},{"id":"north-mini-code-free","name":"North Mini Code Free","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-06-10"},{"id":"qwen3.5-plus","name":"Qwen3.5 Plus","created_by":"qwen","source":"official","family":"qwen3.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.2,"output":1.2,"cached_input":0.02,"cache_write":0.25},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-16","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","context_window":1000000,"max_output_tokens":64000,"license":"proprietary","open_weight":false},{"id":"qwen3.6-plus-free","name":"Qwen3.6 Plus Free","created_by":"qwen","source":"official","family":"qwen3.6","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-01"},{"id":"qwen3.6-plus","name":"Qwen3.6 Plus","created_by":"qwen","source":"official","family":"qwen3.6","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3,"cached_input":0.05,"cache_write":0.625},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-16","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks.","context_window":1000000,"max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"qwen3.7-max","name":"Qwen3.7 Max","created_by":"qwen","source":"official","family":"qwen3.7","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":2.5,"output":7.5,"cached_input":0.5,"cache_write":3.125},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02","context_window":1000000,"license":"apache-2.0","open_weight":true},{"id":"qwen3.7-plus","name":"Qwen3.7 Plus","created_by":"qwen","source":"official","family":"qwen3.7","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.4,"output":1.6,"cached_input":0.04,"cache_write":0.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10","context_window":1000000,"license":"apache-2.0","open_weight":true},{"id":"ring-2.6-1t-free","name":"Ring 2.6 1T","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-05-14"}]};
+export const provider: ProviderWithModels = {"id":"opencode","name":"OpenCode Zen","region":"US","headquarters":"San Francisco, CA","founded":2025,"url":"https://opencode.ai","api_url":"https://opencode.ai/zen/v1","docs_url":"https://opencode.ai/docs/zen","pricing_url":"https://opencode.ai/docs/zen","description":"AI coding assistant platform providing access to frontier models for software development.","type":"cloud","openai_compatible":true,"free_tier":true,"twitter_url":"https://x.com/anomalyco","discord_url":"https://discord.com/invite/opencode","terms_url":"https://opencode.ai/legal/terms-of-service","support_url":"https://github.com/opencode-ai/opencode","icon":"","models":[{"id":"big-pickle","name":"Big Pickle","created_by":"opencode","source":"official","last_updated":"2026-03-21","status":"active","capabilities":{"streaming":true}},{"id":"claude-3-5-haiku","name":"Claude Haiku 3.5","created_by":"anthropic","source":"official","last_updated":"2026-04-26","family":"claude-haiku","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true},"pricing":{"input":0.8,"output":4,"cached_input":0.08,"cache_write":1},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-02-16","license":"proprietary","open_weight":false},{"id":"claude-fable-5","name":"Claude Fable 5","created_by":"anthropic","source":"official","family":"claude-fable","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":50,"cached_input":1,"cache_write":12.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","tagline":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","context_window":1000000,"max_output_tokens":128000,"license":"proprietary","open_weight":false},{"id":"claude-haiku-4-5","name":"Claude Haiku 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":1,"output":5,"cached_input":0.1,"cache_write":1.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The fastest model with near-frontier intelligence","tagline":"The fastest model with near-frontier intelligence","context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-02","training_data_cutoff":"2025-07","license":"proprietary","open_weight":false},{"id":"claude-opus-4-1","name":"Claude Opus 4.1","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":15,"output":75,"cached_input":1.5,"cache_write":18.75},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":32000,"knowledge_cutoff":"2025-01","training_data_cutoff":"2025-03","license":"proprietary","open_weight":false},{"id":"claude-opus-4-5","name":"Claude Opus 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"claude-opus-4-6","name":"Claude Opus 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The most intelligent broadly available model for agents and coding","tagline":"The most intelligent model for building agents and coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"claude-opus-4-7","name":"Claude Opus 4.7","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-17","description":"Our most capable generally available model for complex reasoning and agentic coding","tagline":"Our most capable generally available model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"claude-opus-4-8","name":"Claude Opus 4.8","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5,"cache_write":6.25},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-29","description":"Anthropic's most capable Opus-tier model for complex reasoning and agentic coding","tagline":"Anthropic's most capable model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4-5","name":"Claude Sonnet 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-01","training_data_cutoff":"2025-07","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.3,"cache_write":3.75},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The best combination of speed and intelligence","tagline":"The best combination of speed and intelligence","context_window":1000000,"max_output_tokens":64000,"knowledge_cutoff":"2025-08","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4","name":"Claude Sonnet 4","created_by":"anthropic","source":"official","last_updated":"2026-04-26","family":"claude-sonnet","status":"deprecated","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-06-15"},{"id":"deepseek-v4-flash-free","name":"DeepSeek V4 Flash Free","created_by":"unknown","source":"official","family":"deepseek","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-14"},{"id":"deepseek-v4-flash","name":"DeepSeek V4 Flash","created_by":"unknown","source":"official","family":"deepseek","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.14,"output":0.28,"cached_input":0.028},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"deepseek-v4-pro","name":"DeepSeek V4 Pro","created_by":"unknown","source":"official","family":"deepseek","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.74,"output":3.48,"cached_input":0.145},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"gemini-3-flash","name":"Gemini 3 Flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","status":"active","capabilities":{"streaming":true},"pricing":{"input":0.5,"output":3,"cached_input":0.05},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"gemini-3.1-pro","name":"Gemini 3.1 Pro","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","status":"active","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"gemini-3.5-flash","name":"Gemini 3.5 Flash","created_by":"google","source":"official","family":"gemini-3.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":1.5,"output":9,"cached_input":0.15},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-23","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.5 Flash Gemini 3.5 Flash provides sustained frontier-level intelligence optimized for real-world tasks at a higher speed and lower cost. Designed for the agentic era, it excels at sub-agent deployment, multi-step workflows, and long-horizon tasks at scale. This model is particularly effective for rapid agentic loops involving complex coding cycles and iterations.","tagline":"The Interactions API is now generally available.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"glm-5.1","name":"GLM 5.1","created_by":"zhipu","source":"official","family":"glm-5.1","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.4,"output":4.4,"cached_input":0.26},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-08"},{"id":"glm-5.2","name":"GLM 5.2","created_by":"zhipu","source":"official","family":"glm-5.2","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.4,"output":4.4,"cached_input":0.26},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29"},{"id":"glm-5","name":"GLM 5","created_by":"zhipu","source":"official","last_updated":"2026-04-26","status":"deprecated","capabilities":{"streaming":true},"pricing":{"input":1,"output":3.2,"cached_input":0.2},"model_type":"chat","family":"glm-5","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-05-14"},{"id":"gpt-5-codex","name":"GPT 5 Codex","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5 optimized for agentic coding in Codex","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5-nano","name":"GPT 5 Nano","created_by":"openai","source":"official","last_updated":"2026-05-07","family":"gpt-5","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.4,"cached_input":0.005},"description":"GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for","tagline":"Fastest, most cost-efficient version of GPT-5","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"gpt-5.1-codex-max","name":"GPT 5.1 Codex Max","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.1","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT‑5.1-Codex-Max is purpose-built for agentic coding.","tagline":"A version of GPT-5.1-codex optimized for long running tasks.","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.1-codex-mini","name":"GPT 5.1 Codex Mini","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.1","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5.1-Codex mini is a smaller, more cost-effective, less-capable version of GPT-5.1-Codex.","tagline":"Smaller, more cost-effective, less-capable version of GPT-5.1-Codex","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.1-codex","name":"GPT 5.1 Codex","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.1","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5.1-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5.1 optimized for agentic coding in Codex.","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.1","name":"GPT 5.1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1 is our flagship model for coding and agentic tasks with configurable reasoning and non-reasoning effort.","tagline":"The best model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"gpt-5.2-codex","name":"GPT 5.2 Codex","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-5.2","status":"deprecated","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"deprecation_date":"2026-07-23","description":"GPT-5.2-Codex is an upgraded version of GPT-5.2 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.2-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.2-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"Our most intelligent coding model optimized for long-horizon, agentic coding tasks.","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.2","name":"GPT 5.2","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.3-codex-spark","name":"GPT 5.3 Codex Spark","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","status":"active","capabilities":{"streaming":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"gpt-5.3-codex","name":"GPT 5.3 Codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.3-Codex is optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.3-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.3-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"The most capable agentic coding model to date.","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4-mini","name":"GPT 5.4 Mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5,"cached_input":0.075},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 mini brings the strengths of GPT-5.4 to a faster, more efficient","tagline":"Our strongest mini model yet for coding, computer use, and subagents","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4-nano","name":"GPT 5.4 Nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25,"cached_input":0.02},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 nano is designed for tasks where speed and cost matter most like","tagline":"Our cheapest GPT-5.4-class model for simple high-volume tasks","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4-pro","name":"GPT 5.4 Pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false},"pricing":{"input":30,"output":180,"cached_input":30},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.4","name":"GPT 5.4","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15,"cached_input":0.25},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 is our frontier model for complex professional work.","tagline":"A more affordable model for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"gpt-5.5-pro","name":"GPT 5.5 Pro","created_by":"openai","source":"official","family":"gpt-5.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"reasoning":true,"batch":true},"pricing":{"input":30,"output":180,"cached_input":30},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"gpt-5.5","name":"GPT 5.5","created_by":"openai","source":"official","family":"gpt-5.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"reasoning":true,"batch":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25","description":"GPT-5.5 is our newest frontier model for the most complex professional work.","tagline":"A new class of intelligence for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"gpt-5","name":"GPT 5","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.07,"output":8.5,"cached_input":0.107},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"grok-build-0.1","name":"Grok Build 0.1","created_by":"unknown","source":"official","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"output":2,"cached_input":0.2,"input":1},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02"},{"id":"hy3-preview-free","name":"Hy3 Preview Free","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-04-24"},{"id":"kimi-k2.5","name":"Kimi K2.5","created_by":"moonshot","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":3,"cached_input":0.1},"family":"kimi-k2.5","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","tagline":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","context_window":256000,"max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"kimi-k2.6","name":"Kimi K2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","status":"active","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.95,"output":4,"cached_input":0.16},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21"},{"id":"ling-2.6-flash","name":"Ling 2.6 Flash","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-04-22"},{"id":"mimo-v2-omni-free","name":"MiMo V2 Omni Free","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"model_type":"chat","family":"mimo","modalities":{"input":["text"],"output":["text"]}},{"id":"mimo-v2-pro-free","name":"MiMo V2 Pro Free","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"model_type":"chat","family":"mimo","modalities":{"input":["text"],"output":["text"]}},{"id":"mimo-v2.5-free","name":"MiMo-V2.5 Free","created_by":"xiaomi","source":"official","family":"mimo","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-29"},{"id":"minimax-m2.5-free","name":"MiniMax M2.5 Free","created_by":"minimax","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"minimax-m2.5","name":"MiniMax M2.5","created_by":"minimax","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06,"cache_write":0.375},"family":"minimax","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Peak Performance. Ultimate Value. Master the Complex","tagline":"Peak Performance.","context_window":204800,"license":"mit","open_weight":true},{"id":"minimax-m2.7","name":"MiniMax M2.7","created_by":"minimax","source":"official","family":"minimax","status":"active","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06,"cache_write":0.375},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","context_window":204800,"license":"proprietary","open_weight":false},{"id":"nemotron-3-super-free","name":"Nemotron 3 Super Free","created_by":"nvidia","source":"official","last_updated":"2026-03-23","status":"active","capabilities":{"streaming":true},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"nemotron-3-ultra-free","name":"Nemotron 3 Ultra Free","created_by":"nvidia","source":"official","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10"},{"id":"north-mini-code-free","name":"North Mini Code Free","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-06-10"},{"id":"qwen3.5-plus","name":"Qwen3.5 Plus","created_by":"qwen","source":"official","family":"qwen3.5","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.2,"output":1.2,"cached_input":0.02,"cache_write":0.25},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-16","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","context_window":1000000,"max_output_tokens":64000,"license":"proprietary","open_weight":false},{"id":"qwen3.6-plus-free","name":"Qwen3.6 Plus Free","created_by":"qwen","source":"official","family":"qwen3.6","status":"active","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-01"},{"id":"qwen3.6-plus","name":"Qwen3.6 Plus","created_by":"qwen","source":"official","family":"qwen3.6","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3,"cached_input":0.05,"cache_write":0.625},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-16","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks.","context_window":1000000,"max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"qwen3.7-max","name":"Qwen3.7 Max","created_by":"qwen","source":"official","family":"qwen3.7","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":2.5,"output":7.5,"cached_input":0.5,"cache_write":3.125},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02","context_window":1000000,"license":"apache-2.0","open_weight":true},{"id":"qwen3.7-plus","name":"Qwen3.7 Plus","created_by":"qwen","source":"official","family":"qwen3.7","status":"active","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.4,"output":1.6,"cached_input":0.04,"cache_write":0.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10","context_window":1000000,"license":"apache-2.0","open_weight":true},{"id":"ring-2.6-1t-free","name":"Ring 2.6 1T","created_by":"unknown","source":"official","status":"active","capabilities":{"streaming":true},"last_updated":"2026-05-14"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/openrouter.ts b/packages/npm/src/providers/openrouter.ts
index e967cb3eb..ed37ed98d 100644
--- a/packages/npm/src/providers/openrouter.ts
+++ b/packages/npm/src/providers/openrouter.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"openrouter","name":"OpenRouter","region":"US","headquarters":"New York, NY","founded":2023,"url":"https://openrouter.ai","api_url":"https://openrouter.ai/api/v1","docs_url":"https://openrouter.ai/docs","pricing_url":"https://openrouter.ai/models","description":"Unified API gateway providing access to 200+ models from multiple providers with standardized pricing.","type":"aggregator","playground_url":"https://openrouter.ai/playground","status_url":"https://status.openrouter.ai","openai_compatible":true,"free_tier":true,"github_url":"https://github.com/OpenRouterTeam","models_url":"https://openrouter.ai/models","twitter_url":"https://x.com/OpenRouterAI","discord_url":"https://discord.com/invite/openrouter","terms_url":"https://openrouter.ai/terms","support_url":"https://openrouter.ai/support","icon":"","models":[{"id":"~anthropic/claude-fable-latest","name":"Anthropic: Claude Fable Latest","created_by":"~anthropic","source":"official","family":"claude-fable","description":"This model always redirects to the latest model in the Claude Fable family.","release_date":"2026-06-09","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Claude Fable family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":10,"output":50,"cached_input":1},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"~anthropic/claude-haiku-latest","name":"Anthropic Claude Haiku Latest","created_by":"~anthropic","source":"official","family":"claude-haiku","description":"This model always redirects to the latest model in the Anthropic Claude Haiku family.","release_date":"2026-04-27","context_window":200000,"max_output_tokens":64000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Anthropic Claude Haiku family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1,"output":5,"cached_input":0.1},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"~anthropic/claude-opus-latest","name":"Anthropic: Claude Opus Latest","created_by":"~anthropic","source":"official","family":"claude-opus","description":"This model always redirects to the latest model in the Claude Opus family.","release_date":"2026-04-21","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Claude Opus family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":25,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-04-22"},{"id":"~anthropic/claude-sonnet-latest","name":"Anthropic Claude Sonnet Latest","created_by":"~anthropic","source":"official","family":"claude-sonnet","description":"This model always redirects to the latest model in the Anthropic Claude Sonnet family.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Anthropic Claude Sonnet family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":3,"output":15,"cached_input":0.3},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"~google/gemini-flash-latest","name":"Google Gemini Flash Latest","created_by":"~google","source":"official","description":"This model always redirects to the latest model in the Google Gemini Flash family.","release_date":"2026-04-27","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"This model always redirects to the latest model in the Google Gemini Flash family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"pricing":{"input":1.5,"output":9,"cached_input":0.15},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"~google/gemini-pro-latest","name":"Google Gemini Pro Latest","created_by":"~google","source":"official","description":"This model always redirects to the latest model in the Google Gemini Pro family.","release_date":"2026-04-27","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"This model always redirects to the latest model in the Google Gemini Pro family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["audio","image","text","video"],"output":["text"]},"pricing":{"input":2,"output":12,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"~moonshotai/kimi-latest","name":"MoonshotAI Kimi Latest","created_by":"~moonshotai","source":"official","family":"kimi","description":"This model always redirects to the latest model in the MoonshotAI Kimi family.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"This model always redirects to the latest model in the MoonshotAI Kimi family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.67,"output":3.5,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"~openai/gpt-latest","name":"OpenAI GPT Latest","created_by":"~openai","source":"official","description":"This model always redirects to the latest model in the OpenAI GPT family.","release_date":"2026-04-27","context_window":1050000,"max_output_tokens":128000,"tagline":"This model always redirects to the latest model in the OpenAI GPT family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"~openai/gpt-mini-latest","name":"OpenAI GPT Mini Latest","created_by":"~openai","source":"official","description":"This model always redirects to the latest model in the OpenAI GPT Mini family.","release_date":"2026-04-27","context_window":400000,"max_output_tokens":128000,"tagline":"This model always redirects to the latest model in the OpenAI GPT Mini family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.75,"output":4.5,"cached_input":0.075},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"ai21/jamba-large-1.7","name":"AI21: Jamba Large 1.7","created_by":"ai21","source":"official","last_updated":"2026-03-24","description":"Jamba Large 1.7 is the latest model in the Jamba open family, offering improvements in grounding, instruction-following, and overall efficiency.","context_window":256000,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":2,"output":8},"model_type":"chat","release_date":"2025-08-08","tagline":"Jamba Large 1.7 is the latest model in the Jamba open family, offering improvements in grounding, instruction-following, and overall efficiency.","tools":["function_calling"]},{"id":"aion-labs/aion-1.0-mini","name":"AionLabs: Aion-1.0-Mini","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-1.0-Mini 32B parameter model is a distilled version of the DeepSeek-R1 model, designed for strong performance in reasoning domains such as mathematics, coding, and logic.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.7,"output":1.4},"release_date":"2025-02-04","tagline":"Aion-1.0-Mini 32B parameter model is a distilled version of the DeepSeek-R1 model, designed for strong performance in reasoning domains such as mathematics, coding, and logic."},{"id":"aion-labs/aion-1.0","name":"AionLabs: Aion-1.0","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-1.0 is a multi-model system designed for high performance across various tasks, including reasoning and coding.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":4,"output":8},"release_date":"2025-02-04","tagline":"Aion-1.0 is a multi-model system designed for high performance across various tasks, including reasoning and coding."},{"id":"aion-labs/aion-2.0","name":"AionLabs: Aion-2.0","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.8,"output":1.6,"cached_input":0.2},"release_date":"2026-02-23","tagline":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling."},{"id":"aion-labs/aion-rp-llama-3.1-8b","name":"AionLabs: Aion-RP 1.0 (8B)","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.6},"parameters":8,"release_date":"2025-02-04","tagline":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses."},{"id":"alfredpros/codellama-7b-instruct-solidity","name":"AlfredPros: CodeLLaMa 7B Instruct Solidity","created_by":"alfredpros","source":"official","last_updated":"2026-03-23","description":"A finetuned 7 billion parameters Code LLaMA - Instruct model to generate Solidity smart contract using 4-bit QLoRA finetuning provided by PEFT library.","context_window":4096,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.2},"family":"codellama","model_type":"code","parameters":7,"release_date":"2025-04-14","tagline":"A finetuned 7 billion parameters Code LLaMA - Instruct model to generate Solidity smart contract using 4-bit QLoRA finetuning provided by PEFT library."},{"id":"alibaba/tongyi-deepresearch-30b-a3b","name":"Tongyi DeepResearch 30B A3B","created_by":"alibaba","source":"official","last_updated":"2026-03-24","description":"Tongyi DeepResearch is an agentic large language model developed by Tongyi Lab, with 30 billion total parameters activating only 3 billion per token.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.09,"output":0.45,"cached_input":0.09},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-09-18","tagline":"Tongyi DeepResearch is an agentic large language model developed by Tongyi Lab, with 30 billion total parameters activating only 3 billion per token.","tools":["function_calling"]},{"id":"allenai/molmo-2-8b","name":"AllenAI: Molmo2 8B","created_by":"allenai","source":"official","last_updated":"2026-03-21","description":"Molmo2-8B is an open vision-language model developed by the Allen Institute for AI (Ai2) as part of the Molmo2 family, supporting image, video, and multi-image understanding and grounding.","status":"deprecated","deprecation_date":"2026-03-23","context_window":36864,"max_output_tokens":36864,"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"parameters":8},{"id":"allenai/olmo-2-0325-32b-instruct","name":"AllenAI: Olmo 2 32B Instruct","created_by":"allenai","source":"official","last_updated":"2026-03-23","description":"OLMo-2 32B Instruct is a supervised instruction-finetuned variant of the OLMo-2 32B March 2025 base model.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.2},"parameters":32,"release_date":"2025-03-14","tagline":"OLMo-2 32B Instruct is a supervised instruction-finetuned variant of the OLMo-2 32B March 2025 base model."},{"id":"allenai/olmo-3-32b-think","name":"AllenAI: Olmo 3 32B Think","created_by":"allenai","source":"official","last_updated":"2026-03-24","description":"Olmo 3 32B Think is a large-scale, 32-billion-parameter model purpose-built for deep reasoning, complex logic chains and advanced instruction-following scenarios.","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.15,"output":0.5},"parameters":32,"release_date":"2025-11-21","tagline":"Olmo 3 32B Think is a large-scale, 32-billion-parameter model purpose-built for deep reasoning, complex logic chains and advanced instruction-following scenarios."},{"id":"allenai/olmo-3-7b-instruct","name":"AllenAI: Olmo 3 7B Instruct","created_by":"allenai","source":"official","last_updated":"2026-03-21","description":"Olmo 3 7B Instruct is a supervised instruction-fine-tuned variant of the Olmo 3 7B base model, optimized for instruction-following, question-answering, and natural conversational dialogue.","status":"deprecated","deprecation_date":"2026-03-23","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.2},"parameters":7},{"id":"allenai/olmo-3-7b-think","name":"AllenAI: Olmo 3 7B Think","created_by":"allenai","source":"official","last_updated":"2026-03-21","description":"Olmo 3 7B Think is a research-oriented language model in the Olmo family designed for advanced reasoning and instruction-driven tasks.","status":"deprecated","deprecation_date":"2026-03-23","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.12,"output":0.2},"parameters":7},{"id":"allenai/olmo-3.1-32b-instruct","name":"AllenAI: Olmo 3.1 32B Instruct","created_by":"allenai","source":"official","last_updated":"2026-05-01","description":"Olmo 3.1 32B Instruct is a large-scale, 32-billion-parameter instruction-tuned language model engineered for high-performance conversational AI, multi-turn dialogue, and practical instruction following.","context_window":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.6},"parameters":32,"release_date":"2026-01-06","tagline":"Olmo 3.1 32B Instruct is a large-scale, 32-billion-parameter instruction-tuned language model engineered for high-performance conversational AI, multi-turn dialogue, and practical instruction followin","tools":["function_calling"],"max_output_tokens":16384,"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"allenai/olmo-3.1-32b-think","name":"AllenAI: Olmo 3.1 32B Think","created_by":"allenai","source":"official","last_updated":"2026-03-24","description":"Olmo 3.1 32B Think is a large-scale, 32-billion-parameter model designed for deep reasoning, complex multi-step logic, and advanced instruction following.","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.15,"output":0.5},"parameters":32,"status":"deprecated","release_date":"2025-12-16","deprecation_date":"2026-04-06","tagline":"Olmo 3.1 32B Think is a large-scale, 32-billion-parameter model designed for deep reasoning, complex multi-step logic, and advanced instruction following."},{"id":"alpindale/goliath-120b","name":"Goliath 120B","created_by":"alpindale","source":"official","last_updated":"2026-03-24","description":"A large LLM created by combining two fine-tuned Llama 70B models into one 120B model.","context_window":6144,"max_output_tokens":1024,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":3.75,"output":7.5},"parameters":120,"release_date":"2023-11-10","tagline":"A large LLM created by combining two fine-tuned Llama 70B models into one 120B model."},{"id":"amazon/nova-2-lite-v1","name":"Amazon: Nova 2 Lite","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text.","context_window":1000000,"max_output_tokens":65535,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","release_date":"2025-12-02","tagline":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text.","tools":["function_calling"]},{"id":"amazon/nova-lite-v1","name":"Amazon: Nova Lite 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output.","context_window":300000,"max_output_tokens":5120,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.06,"output":0.24},"model_type":"chat","release_date":"2024-12-05","tagline":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output.","tools":["function_calling"]},{"id":"amazon/nova-micro-v1","name":"Amazon: Nova Micro 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost.","context_window":128000,"max_output_tokens":5120,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.035,"output":0.14},"model_type":"chat","release_date":"2024-12-05","tagline":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost.","tools":["function_calling"]},{"id":"amazon/nova-premier-v1","name":"Amazon: Nova Premier 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","context_window":1000000,"max_output_tokens":32000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":2.5,"output":12.5,"cached_input":0.625},"model_type":"chat","release_date":"2025-10-31","tagline":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","tools":["function_calling"]},{"id":"amazon/nova-pro-v1","name":"Amazon: Nova Pro 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks.","context_window":300000,"max_output_tokens":5120,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.8,"output":3.2},"model_type":"chat","release_date":"2024-12-05","tagline":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks.","tools":["function_calling"]},{"id":"anthracite-org/magnum-v4-72b","name":"Magnum v4 72B","created_by":"anthracite-org","source":"official","last_updated":"2026-05-17","description":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus).","context_window":32768,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":3,"output":5},"parameters":72,"release_date":"2024-10-22","tagline":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthrop"},{"id":"anthropic/claude-3-haiku","name":"Anthropic: Claude 3 Haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","description":"Claude 3 Haiku is Anthropic's fastest and most compact model for\nnear-instant responsiveness. Quick and accurate targeted performance.\n\nSee the launch announcement and benchmark results [here](https:/","context_window":200000,"max_output_tokens":4096,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"streaming":true,"vision":true,"batch":true},"pricing":{"input":0.25,"output":1.25,"cached_input":0.03},"model_type":"chat","release_date":"2024-03-13","tagline":"Claude 3 Haiku is Anthropic's fastest and most compact model for\nnear-instant responsiveness. Quick and accurate targeted performance.\n\nSee the launch announcement and benchmark results [here](https:/","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"anthropic/claude-3.5-haiku","name":"Anthropic: Claude 3.5 Haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","description":"Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use.","context_window":200000,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.8,"output":4,"cached_input":0.08},"model_type":"chat","release_date":"2024-11-04","tagline":"Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use.","tools":["function_calling"]},{"id":"anthropic/claude-3.5-sonnet","name":"Anthropic: Claude 3.5 Sonnet","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","description":"New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices.","context_window":200000,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":6,"output":30,"cached_input":0.6},"model_type":"chat","release_date":"2024-10-22","tagline":"New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices.","tools":["function_calling"]},{"id":"anthropic/claude-3.7-sonnet:thinking","name":"Anthropic: Claude 3.7 Sonnet (thinking)","created_by":"anthropic","source":"official","last_updated":"2026-05-06","family":"claude-sonnet","description":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","status":"deprecated","deprecation_date":"2026-05-11","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-02-24","tagline":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","tools":["function_calling"]},{"id":"anthropic/claude-3.7-sonnet","name":"Anthropic: Claude 3.7 Sonnet","created_by":"anthropic","source":"official","last_updated":"2026-05-06","family":"claude-sonnet","description":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","status":"deprecated","deprecation_date":"2026-05-11","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-02-24","tagline":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","tools":["function_calling"]},{"id":"anthropic/claude-fable-5","name":"Anthropic: Claude Fable 5","created_by":"anthropic","source":"official","family":"claude-fable","description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding.","release_date":"2026-06-09","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":10,"output":50,"cached_input":1},"tools":["function_calling"],"last_updated":"2026-06-18","license":"proprietary","open_weight":false},{"id":"anthropic/claude-haiku-4.5","name":"Anthropic: Claude Haiku 4.5","created_by":"anthropic","source":"official","last_updated":"2026-05-14","family":"claude-haiku","description":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models.","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1,"output":5,"cached_input":0.1},"model_type":"chat","release_date":"2025-10-15","tagline":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.1","name":"Anthropic: Claude Opus 4.1","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-opus","description":"Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks.","context_window":200000,"max_output_tokens":32000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":15,"output":75,"cached_input":1.5},"model_type":"chat","release_date":"2025-08-05","tagline":"Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.5","name":"Anthropic: Claude Opus 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-opus","description":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use.","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","release_date":"2025-11-24","tagline":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.6-fast","name":"Anthropic: Claude Opus 4.6 (Fast)","created_by":"anthropic","source":"official","family":"claude-opus","description":"Fast-mode variant of [Opus 4.6](/anthropic/claude-opus-4.6) - identical capabilities with higher output speed at premium 6x pricing.","release_date":"2026-04-07","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Fast-mode variant of [Opus 4.6](/anthropic/claude-opus-4.6) - identical capabilities with higher output speed at premium 6x pricing.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":30,"output":150,"cached_input":3},"tools":["function_calling"],"last_updated":"2026-05-29","status":"deprecated","deprecation_date":"2026-06-29"},{"id":"anthropic/claude-opus-4.6","name":"Anthropic: Claude Opus 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-opus","description":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks.","context_window":1000000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","release_date":"2026-02-04","tagline":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.7-fast","name":"Anthropic: Claude Opus 4.7 (Fast)","created_by":"anthropic","source":"official","family":"claude-opus","description":"Fast-mode variant of [Opus 4.7](/anthropic/claude-opus-4.7) - identical capabilities with higher output speed at premium 6x pricing.","release_date":"2026-05-12","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Fast-mode variant of [Opus 4.7](/anthropic/claude-opus-4.7) - identical capabilities with higher output speed at premium 6x pricing.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":30,"output":150,"cached_input":3},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"anthropic/claude-opus-4.7","name":"Anthropic: Claude Opus 4.7","created_by":"anthropic","source":"official","family":"claude-opus","description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.","release_date":"2026-04-16","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":25,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"anthropic/claude-opus-4.8-fast","name":"Anthropic: Claude Opus 4.8 (Fast)","created_by":"anthropic","source":"official","family":"claude-opus","description":"Fast-mode variant of [Opus 4.8](/anthropic/claude-opus-4.8) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.","release_date":"2026-05-27","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Fast-mode variant of [Opus 4.8](/anthropic/claude-opus-4.8) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":10,"output":50,"cached_input":1},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"anthropic/claude-opus-4.8","name":"Anthropic: Claude Opus 4.8","created_by":"anthropic","source":"official","family":"claude-opus","description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family.","release_date":"2026-05-27","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":25,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"anthropic/claude-opus-4","name":"Anthropic: Claude Opus 4","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","description":"Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows.","context_window":200000,"max_output_tokens":32000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":15,"output":75,"cached_input":1.5},"model_type":"chat","release_date":"2025-05-22","tagline":"Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows.","tools":["function_calling"]},{"id":"anthropic/claude-sonnet-4.5","name":"Anthropic: Claude Sonnet 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-sonnet","description":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows.","context_window":1000000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-09-29","tagline":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows.","tools":["function_calling"]},{"id":"anthropic/claude-sonnet-4.6","name":"Anthropic: Claude Sonnet 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-sonnet","description":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work.","context_window":1000000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2026-02-17","tagline":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work.","tools":["function_calling"]},{"id":"anthropic/claude-sonnet-4","name":"Anthropic: Claude Sonnet 4","created_by":"anthropic","source":"official","last_updated":"2026-04-12","family":"claude-sonnet","description":"Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability.","context_window":1000000,"max_output_tokens":64000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-05-22","tagline":"Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability.","tools":["function_calling"]},{"id":"arcee-ai/coder-large","name":"Arcee AI: Coder Large","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Coder‑Large is a 32 B‑parameter offspring of Qwen 2.5‑Instruct that has been further trained on permissively‑licensed GitHub, CodeSearchNet and synthetic bug‑fix corpora.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":0.8},"release_date":"2025-05-05","tagline":"Coder‑Large is a 32 B‑parameter offspring of Qwen 2.5‑Instruct that has been further trained on permissively‑licensed GitHub, CodeSearchNet and synthetic bug‑fix corpora."},{"id":"arcee-ai/maestro-reasoning","name":"Arcee AI: Maestro Reasoning","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Maestro Reasoning is Arcee's flagship analysis model: a 32 B‑parameter derivative of Qwen 2.5‑32 B tuned with DPO and chain‑of‑thought RL for step‑by‑step logic.","context_window":131072,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.9,"output":3.3},"model_type":"reasoning","release_date":"2025-05-05","tagline":"Maestro Reasoning is Arcee's flagship analysis model: a 32 B‑parameter derivative of Qwen 2.5‑32 B tuned with DPO and chain‑of‑thought RL for step‑by‑step logic."},{"id":"arcee-ai/spotlight","name":"Arcee AI: Spotlight","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Spotlight is a 7‑billion‑parameter vision‑language model derived from Qwen 2.5‑VL and fine‑tuned by Arcee AI for tight image‑text grounding tasks.","context_window":131072,"max_output_tokens":65537,"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.18,"output":0.18},"release_date":"2025-05-05","tagline":"Spotlight is a 7‑billion‑parameter vision‑language model derived from Qwen 2.5‑VL and fine‑tuned by Arcee AI for tight image‑text grounding tasks."},{"id":"arcee-ai/trinity-large-preview:free","name":"Arcee AI: Trinity Large Preview (free)","created_by":"arcee-ai","source":"official","last_updated":"2026-04-10","description":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","context_window":131000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"release_date":"2026-01-27","tagline":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-22"},{"id":"arcee-ai/trinity-large-preview","name":"Arcee AI: Trinity Large Preview","created_by":"arcee-ai","source":"official","description":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","release_date":"2026-01-27","context_window":131000,"tagline":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.45},"tools":["function_calling"],"last_updated":"2026-04-24"},{"id":"arcee-ai/trinity-large-thinking:free","name":"Arcee AI: Trinity Large Thinking (free)","created_by":"arcee-ai","source":"official","description":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","release_date":"2026-04-01","context_window":262144,"max_output_tokens":80000,"tagline":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"arcee-ai/trinity-large-thinking","name":"Arcee AI: Trinity Large Thinking","created_by":"arcee-ai","source":"official","description":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","release_date":"2026-04-01","context_window":262144,"max_output_tokens":262144,"tagline":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.22,"output":0.85,"cached_input":0.06},"tools":["function_calling"],"last_updated":"2026-04-14"},{"id":"arcee-ai/trinity-mini:free","name":"Arcee AI: Trinity Mini (free)","created_by":"arcee-ai","source":"official","last_updated":"2026-04-03","description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"release_date":"2025-12-01","tagline":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-10"},{"id":"arcee-ai/trinity-mini","name":"Arcee AI: Trinity Mini","created_by":"arcee-ai","source":"official","last_updated":"2026-03-24","description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.045,"output":0.15},"release_date":"2025-12-01","tagline":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","tools":["function_calling"]},{"id":"arcee-ai/virtuoso-large","name":"Arcee AI: Virtuoso Large","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Virtuoso‑Large is Arcee's top‑tier general‑purpose LLM at 72 B parameters, tuned to tackle cross‑domain reasoning, creative writing and enterprise QA.","context_window":131072,"max_output_tokens":64000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.75,"output":1.2},"release_date":"2025-05-05","tagline":"Virtuoso‑Large is Arcee's top‑tier general‑purpose LLM at 72 B parameters, tuned to tackle cross‑domain reasoning, creative writing and enterprise QA.","tools":["function_calling"]},{"id":"baidu/cobuddy:free","name":"Baidu Qianfan: CoBuddy (free)","created_by":"baidu","source":"official","description":"CoBuddy is a code generation model from Baidu, optimized for coding tasks and AI Agent workflows.","release_date":"2026-05-06","context_window":131072,"max_output_tokens":65536,"tagline":"CoBuddy is a code generation model from Baidu, optimized for coding tasks and AI Agent workflows.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-06"},{"id":"baidu/ernie-4.5-21b-a3b-thinking","name":"Baidu: ERNIE 4.5 21B A3B Thinking","created_by":"baidu","source":"official","last_updated":"2026-03-23","description":"ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science, coding, text generation, ","context_window":131072,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.07,"output":0.28},"parameters":21,"active_parameters":3,"release_date":"2025-10-09","tagline":"ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science, coding, text generation, "},{"id":"baidu/ernie-4.5-21b-a3b","name":"Baidu: ERNIE 4.5 21B A3B","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"A sophisticated text-based Mixture-of-Experts (MoE) model featuring 21B total parameters with 3B activated per token, delivering exceptional multimodal understanding and generation through heterogeneous MoE structures and modality-isolated routing.","context_window":131072,"max_output_tokens":8000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.07,"output":0.28},"parameters":21,"active_parameters":3,"release_date":"2025-08-12","tagline":"A sophisticated text-based Mixture-of-Experts (MoE) model featuring 21B total parameters with 3B activated per token, delivering exceptional multimodal understanding and generation through heterogeneo","tools":["function_calling"]},{"id":"baidu/ernie-4.5-300b-a47b","name":"Baidu: ERNIE 4.5 300B A47B ","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"ERNIE-4.5-300B-A47B is a 300B parameter Mixture-of-Experts (MoE) language model developed by Baidu as part of the ERNIE 4.5 series.","context_window":131072,"max_output_tokens":12000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.28,"output":1.1},"parameters":300,"active_parameters":47,"release_date":"2025-06-30","tagline":"ERNIE-4.5-300B-A47B is a 300B parameter Mixture-of-Experts (MoE) language model developed by Baidu as part of the ERNIE 4.5 series."},{"id":"baidu/ernie-4.5-vl-28b-a3b","name":"Baidu: ERNIE 4.5 VL 28B A3B","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"A powerful multimodal Mixture-of-Experts chat model featuring 28B total parameters with 3B activated per token, delivering exceptional text and vision understanding through its innovative heterogeneou","context_window":131072,"max_output_tokens":8000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":0.14,"output":0.56},"parameters":28,"active_parameters":3,"release_date":"2025-08-12","tagline":"A powerful multimodal Mixture-of-Experts chat model featuring 28B total parameters with 3B activated per token, delivering exceptional text and vision understanding through its innovative heterogeneou","tools":["function_calling"]},{"id":"baidu/ernie-4.5-vl-424b-a47b","name":"Baidu: ERNIE 4.5 VL 424B A47B ","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active per token.","context_window":131072,"max_output_tokens":16000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.42,"output":1.25},"parameters":424,"active_parameters":47,"release_date":"2025-06-30","tagline":"ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active per token."},{"id":"baidu/qianfan-ocr-fast:free","name":"Baidu: Qianfan-OCR-Fast (free)","created_by":"baidu","source":"official","description":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","release_date":"2026-04-20","context_window":65536,"max_output_tokens":28672,"tagline":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","capabilities":{"reasoning":true},"modalities":{"input":["image","text"],"output":["text"]},"last_updated":"2026-04-24"},{"id":"baidu/qianfan-ocr-fast","name":"Baidu: Qianfan-OCR-Fast","created_by":"baidu","source":"official","description":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","release_date":"2026-04-20","context_window":65536,"max_output_tokens":28672,"tagline":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","capabilities":{"reasoning":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.68,"output":2.81},"last_updated":"2026-05-23","status":"deprecated","deprecation_date":"2026-05-28"},{"id":"bytedance-seed/seed-1.6-flash","name":"ByteDance Seed: Seed 1.6 Flash","created_by":"bytedance-seed","source":"official","last_updated":"2026-03-24","description":"Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.075,"output":0.3},"release_date":"2025-12-23","tagline":"Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding.","tools":["function_calling"]},{"id":"bytedance-seed/seed-1.6","name":"ByteDance Seed: Seed 1.6","created_by":"bytedance-seed","source":"official","last_updated":"2026-03-24","description":"Seed 1.6 is a general-purpose model released by the ByteDance Seed team.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.25,"output":2},"release_date":"2025-12-23","tagline":"Seed 1.6 is a general-purpose model released by the ByteDance Seed team.","tools":["function_calling"]},{"id":"bytedance-seed/seed-2.0-lite","name":"ByteDance Seed: Seed-2.0-Lite","created_by":"bytedance-seed","source":"official","last_updated":"2026-04-07","description":"Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice f","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.25,"output":2},"release_date":"2026-03-10","tagline":"Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice f","tools":["function_calling"]},{"id":"bytedance-seed/seed-2.0-mini","name":"ByteDance Seed: Seed-2.0-Mini","created_by":"bytedance-seed","source":"official","last_updated":"2026-03-24","description":"Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment.","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"release_date":"2026-02-26","tagline":"Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment.","tools":["function_calling"]},{"id":"bytedance/ui-tars-1.5-7b","name":"ByteDance: UI-TARS 7B ","created_by":"bytedance","source":"official","last_updated":"2026-06-18","description":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games.","context_window":128000,"max_output_tokens":2048,"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.1,"output":0.2,"cached_input":0.1},"parameters":7,"release_date":"2025-07-22","tagline":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"cognitivecomputations/dolphin-mistral-24b-venice-edition:free","name":"Venice: Uncensored (free)","created_by":"cognitivecomputations","source":"official","last_updated":"2026-03-24","description":"Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"parameters":24,"release_date":"2025-07-09","tagline":"Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai."},{"id":"cohere/command-a","name":"Cohere: Command A","created_by":"cohere","source":"official","last_updated":"2026-03-24","description":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.","context_window":256000,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":2.5,"output":10},"family":"command-a","model_type":"chat","release_date":"2025-03-13","tagline":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases."},{"id":"cohere/command-r-08-2024","name":"Cohere: Command R (08-2024)","created_by":"cohere","source":"official","last_updated":"2026-03-24","description":"command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use.","context_window":128000,"max_output_tokens":4000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.6},"family":"command-r","model_type":"chat","release_date":"2024-08-30","tagline":"command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use.","tools":["function_calling"],"license":"cc-by-nc-4.0","open_weight":true},{"id":"cohere/command-r-plus-08-2024","name":"Cohere: Command R+ (08-2024)","created_by":"cohere","source":"official","last_updated":"2026-04-07","description":"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while ","context_window":128000,"max_output_tokens":4000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":2.5,"output":10},"family":"command-r-plus","model_type":"chat","release_date":"2024-08-30","tagline":"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while ","tools":["function_calling"],"license":"cc-by-nc-4.0","open_weight":true},{"id":"cohere/command-r7b-12-2024","name":"Cohere: Command R7B (12-2024)","created_by":"cohere","source":"official","last_updated":"2026-03-24","description":"Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024.","context_window":128000,"max_output_tokens":4000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.038,"output":0.15},"family":"command-r","model_type":"chat","release_date":"2024-12-14","tagline":"Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024.","license":"cc-by-nc-4.0","open_weight":true},{"id":"cohere/north-mini-code:free","name":"Cohere: North Mini Code (free)","created_by":"cohere","source":"official","description":"North Mini Code is Cohere's first agentic coding model and the debut of its North family.","release_date":"2026-06-17","context_window":256000,"max_output_tokens":64000,"tagline":"North Mini Code is Cohere's first agentic coding model and the debut of its North family.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"deepcogito/cogito-v2.1-671b","name":"Deep Cogito: Cogito v2.1 671B","created_by":"deepcogito","source":"official","last_updated":"2026-03-24","description":"Cogito v2.1 671B MoE represents one of the strongest open models globally, matching performance of frontier closed and open models.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.25,"output":1.25},"parameters":671,"release_date":"2025-11-13","tagline":"Cogito v2.1 671B MoE represents one of the strongest open models globally, matching performance of frontier closed and open models.","model_type":"chat"},{"id":"deepseek/deepseek-chat-v3-0324","name":"DeepSeek: DeepSeek V3 0324","created_by":"deepseek","source":"official","last_updated":"2026-04-25","family":"deepseek-chat","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","context_window":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.77,"cached_input":0.135},"release_date":"2025-03-24","tagline":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","tools":["function_calling"],"model_type":"chat","max_output_tokens":16384},{"id":"deepseek/deepseek-chat-v3.1","name":"DeepSeek: DeepSeek V3.1","created_by":"deepseek","source":"official","last_updated":"2026-05-14","family":"deepseek-chat","description":"DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.","context_window":163840,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.21,"output":0.79,"cached_input":0.13},"release_date":"2025-08-21","tagline":"DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.","tools":["function_calling"],"model_type":"chat"},{"id":"deepseek/deepseek-chat","name":"DeepSeek: DeepSeek V3","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek-chat","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","context_window":131072,"max_output_tokens":16000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.8},"release_date":"2024-12-26","tagline":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","tools":["function_calling"],"model_type":"chat","license":"mit","open_weight":true},{"id":"deepseek/deepseek-r1-0528","name":"DeepSeek: R1 0528","created_by":"deepseek","source":"official","last_updated":"2026-04-25","family":"deepseek-r1","description":"May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","context_window":163840,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.5,"output":2.15,"cached_input":0.35},"model_type":"reasoning","release_date":"2025-05-28","tagline":"May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","tools":["function_calling"]},{"id":"deepseek/deepseek-r1-distill-llama-70b","name":"DeepSeek: R1 Distill Llama 70B","created_by":"deepseek","source":"official","last_updated":"2026-06-18","family":"deepseek-r1","description":"DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1).","context_window":128000,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.8,"output":0.8},"model_type":"reasoning","parameters":70,"release_date":"2025-01-23","tagline":"DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1).","status":"deprecated","deprecation_date":"2026-06-11"},{"id":"deepseek/deepseek-r1-distill-qwen-32b","name":"DeepSeek: R1 Distill Qwen 32B","created_by":"deepseek","source":"official","last_updated":"2026-05-17","family":"deepseek-r1","description":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1).","context_window":128000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.29,"output":0.29},"model_type":"reasoning","parameters":32,"release_date":"2025-01-29","tagline":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1)."},{"id":"deepseek/deepseek-r1","name":"DeepSeek: R1","created_by":"deepseek","source":"official","last_updated":"2026-05-17","family":"deepseek-r1","description":"DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","context_window":163840,"max_output_tokens":16000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.7,"output":2.5},"model_type":"reasoning","release_date":"2025-01-20","tagline":"DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","tools":["function_calling"]},{"id":"deepseek/deepseek-v3.1-terminus","name":"DeepSeek: DeepSeek V3.1 Terminus","created_by":"deepseek","source":"official","last_updated":"2026-05-06","family":"deepseek","description":"DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language cons","context_window":163840,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.27,"output":0.95,"cached_input":0.13},"release_date":"2025-09-22","tagline":"DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language cons","tools":["function_calling"],"model_type":"chat"},{"id":"deepseek/deepseek-v3.2-exp","name":"DeepSeek: DeepSeek V3.2 Exp","created_by":"deepseek","source":"official","last_updated":"2026-03-24","family":"deepseek","description":"DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.","context_window":163840,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.27,"output":0.41},"release_date":"2025-09-29","tagline":"DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.","tools":["function_calling"],"model_type":"chat"},{"id":"deepseek/deepseek-v3.2-speciale","name":"DeepSeek: DeepSeek V3.2 Speciale","created_by":"deepseek","source":"official","last_updated":"2026-05-07","family":"deepseek","description":"DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance.","context_window":163840,"max_output_tokens":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.287,"output":0.431,"cached_input":0.058},"release_date":"2025-12-01","tagline":"DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance.","model_type":"chat"},{"id":"deepseek/deepseek-v3.2","name":"DeepSeek: DeepSeek V3.2","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek","description":"DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.229,"output":0.343,"cached_input":0.025},"release_date":"2025-12-01","tagline":"DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.","tools":["function_calling"],"model_type":"chat","max_output_tokens":64000},{"id":"deepseek/deepseek-v4-flash:free","name":"DeepSeek: DeepSeek V4 Flash (free)","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":384000,"model_type":"chat","tagline":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek: DeepSeek V4 Flash","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.18,"cached_input":0.02},"tools":["function_calling"],"last_updated":"2026-06-18","license":"mit","open_weight":true},{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek: DeepSeek V4 Pro","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":384000,"model_type":"chat","tagline":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.435,"output":0.87,"cached_input":0.004},"tools":["function_calling"],"last_updated":"2026-04-26","license":"mit","open_weight":true},{"id":"eleutherai/llemma_7b","name":"EleutherAI: Llemma 7b","created_by":"eleutherai","source":"official","last_updated":"2026-03-23","description":"Llemma 7B is a language model for mathematics.","context_window":4096,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.2},"parameters":7,"release_date":"2025-04-14","tagline":"Llemma 7B is a language model for mathematics."},{"id":"essentialai/rnj-1-instruct","name":"EssentialAI: Rnj 1 Instruct","created_by":"essentialai","source":"official","last_updated":"2026-03-24","description":"Rnj-1 is an 8B-parameter, dense, open-weight model family developed by Essential AI and trained from scratch with a focus on programming, math, and scientific reasoning.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.15,"output":0.15},"release_date":"2025-12-07","tagline":"Rnj-1 is an 8B-parameter, dense, open-weight model family developed by Essential AI and trained from scratch with a focus on programming, math, and scientific reasoning.","tools":["function_calling"],"model_type":"chat"},{"id":"google/gemini-2.0-flash-001","name":"Google: Gemini 2.0 Flash","created_by":"google","source":"official","last_updated":"2026-05-23","family":"gemini-2.0","description":"Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5).","status":"deprecated","deprecation_date":"2026-06-01","context_window":1000000,"max_output_tokens":8192,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.025},"model_type":"chat","release_date":"2025-02-05","tagline":"Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"google/gemini-2.0-flash-lite-001","name":"Google: Gemini 2.0 Flash Lite","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.0","description":"Gemini 2.0 Flash Lite offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemin","status":"deprecated","deprecation_date":"2026-06-01","context_window":1048576,"max_output_tokens":8192,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.075,"output":0.3},"model_type":"chat","release_date":"2025-02-25","tagline":"Gemini 2.0 Flash Lite offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemin","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-image","name":"Google: Nano Banana (Gemini 2.5 Flash Image)","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemini-2.5","description":"Gemini 2.5 Flash Image, a.k.a.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"batch":true,"vision":true},"pricing":{"input":0.3,"output":2.5,"cached_input":0.03},"model_type":"chat","release_date":"2025-10-07","tagline":"Gemini 2.5 Flash Image, a.k.a.","knowledge_cutoff":"2025-06","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite-preview-09-2025","name":"Google: Gemini 2.5 Flash Lite Preview 09-2025","created_by":"google","source":"official","last_updated":"2026-04-03","family":"gemini-2.5","description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.01},"model_type":"chat","release_date":"2025-09-25","tagline":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite","name":"Google: Gemini 2.5 Flash Lite","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.01},"model_type":"chat","release_date":"2025-07-22","tagline":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash","name":"Google: Gemini 2.5 Flash","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["image","text","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.3,"output":2.5,"cached_input":0.03},"model_type":"chat","release_date":"2025-06-17","tagline":"Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-pro-preview-05-06","name":"Google: Gemini 2.5 Pro Preview 05-06","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-05-07","tagline":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-pro-preview","name":"Google: Gemini 2.5 Pro Preview 06-05","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["image","text","audio"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-06-05","tagline":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"]},{"id":"google/gemini-2.5-pro","name":"Google: Gemini 2.5 Pro","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-06-17","tagline":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-flash-preview","name":"Google: Gemini 3 Flash Preview","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemini-3","description":"Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.5,"output":3,"cached_input":0.05},"model_type":"chat","release_date":"2025-12-17","tagline":"Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-image-preview","name":"Google: Nano Banana Pro (Gemini 3 Pro Image Preview)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3","description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","context_window":65536,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2025-11-20","tagline":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-image","name":"Google: Nano Banana Pro (Gemini 3 Pro Image)","created_by":"google","source":"official","family":"gemini-3","description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","release_date":"2026-06-18","context_window":65536,"max_output_tokens":32768,"model_type":"chat","tagline":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["image","text"],"output":["image","text"]},"pricing":{"input":2,"output":12,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-06-18","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-preview","name":"Google: Gemini 3 Pro Preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3","description":"Gemini 3 Pro is Google’s flagship frontier model for high-precision multimodal reasoning, combining strong performance across text, image, video, audio, and code with a 1M-token context window.","status":"deprecated","deprecation_date":"2026-03-26","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2025-11-18","tagline":"Gemini 3 Pro is Google’s flagship frontier model for high-precision multimodal reasoning, combining strong performance across text, image, video, audio, and code with a 1M-token context window.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image-preview","name":"Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)","created_by":"google","source":"official","last_updated":"2026-05-17","family":"gemini-3.1","description":"Gemini 3.1 Flash Image Preview, a.k.a.","context_window":131072,"max_output_tokens":65536,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","release_date":"2026-02-26","tagline":"Gemini 3.1 Flash Image Preview, a.k.a.","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image","name":"Google: Nano Banana 2 (Gemini 3.1 Flash Image)","created_by":"google","source":"official","family":"gemini-3.1","description":"Gemini 3.1 Flash Image, a.k.a.","release_date":"2026-06-18","context_window":131072,"max_output_tokens":65536,"model_type":"chat","tagline":"Gemini 3.1 Flash Image, a.k.a.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"tool_call":true,"batch":true,"vision":true},"modalities":{"input":["image","text"],"output":["image","text"]},"pricing":{"input":0.5,"output":3},"last_updated":"2026-06-18","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite-preview","name":"Google: Gemini 3.1 Flash Lite Preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3.1","description":"Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["text","image","video","audio"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.25,"output":1.5,"cached_input":0.025},"model_type":"chat","release_date":"2026-03-03","tagline":"Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite","name":"Google: Gemini 3.1 Flash Lite","created_by":"google","source":"official","family":"gemini-3.1","description":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads.","release_date":"2026-05-07","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"pricing":{"input":0.25,"output":1.5,"cached_input":0.025},"tools":["function_calling"],"last_updated":"2026-05-10","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-pro-preview-customtools","name":"Google: Gemini 3.1 Pro Preview Custom Tools","created_by":"google","source":"official","last_updated":"2026-05-17","family":"gemini-3.1","description":"Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party...","context_window":1048756,"max_output_tokens":65536,"modalities":{"input":["text","audio","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2026-02-25","tagline":"Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party or user-defined fu","tools":["function_calling"]},{"id":"google/gemini-3.1-pro-preview","name":"Google: Gemini 3.1 Pro Preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3.1","description":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["audio","image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2026-02-19","tagline":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.5-flash","name":"Google: Gemini 3.5 Flash","created_by":"google","source":"official","family":"gemini-3.5","description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed.","release_date":"2026-05-19","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"pricing":{"input":1.5,"output":9,"cached_input":0.15},"tools":["function_calling"],"last_updated":"2026-05-23","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemma-2-27b-it","name":"Google: Gemma 2 27B","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","description":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini).","context_window":8192,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.65,"output":0.65},"model_type":"chat","parameters":27,"release_date":"2024-07-13","tagline":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini)."},{"id":"google/gemma-2-9b-it","name":"Google: Gemma 2 9B","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","description":"Gemma 2 9B by Google is an advanced, open-source language model that sets a new standard for efficiency and performance in its size class.","context_window":8192,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.09},"model_type":"chat","parameters":9,"release_date":"2024-06-28","tagline":"Gemma 2 9B by Google is an advanced, open-source language model that sets a new standard for efficiency and performance in its size class."},{"id":"google/gemma-3-12b-it:free","name":"Google: Gemma 3 12B (free)","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"model_type":"chat","parameters":12,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs."},{"id":"google/gemma-3-12b-it","name":"Google: Gemma 3 12B","created_by":"google","source":"official","last_updated":"2026-06-10","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true},"pricing":{"input":0.05,"output":0.15},"model_type":"chat","parameters":12,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","max_output_tokens":16384,"tools":["function_calling"]},{"id":"google/gemma-3-27b-it:free","name":"Google: Gemma 3 27B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","parameters":27,"release_date":"2025-03-12","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs."},{"id":"google/gemma-3-27b-it","name":"Google: Gemma 3 27B","created_by":"google","source":"official","last_updated":"2026-04-25","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true},"pricing":{"input":0.08,"output":0.16},"model_type":"chat","parameters":27,"release_date":"2025-03-12","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","tools":["function_calling"]},{"id":"google/gemma-3-4b-it:free","name":"Google: Gemma 3 4B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","parameters":4,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs."},{"id":"google/gemma-3-4b-it","name":"Google: Gemma 3 4B","created_by":"google","source":"official","last_updated":"2026-06-10","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.05,"output":0.1},"model_type":"chat","parameters":4,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","max_output_tokens":16384},{"id":"google/gemma-3n-e2b-it:free","name":"Google: Gemma 3n 2B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3n","description":"Gemma 3n E2B IT is a multimodal, instruction-tuned model developed by Google DeepMind, designed to operate efficiently at an effective parameter size of 2B while leveraging a 6B architecture.","context_window":8192,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","release_date":"2025-07-09","tagline":"Gemma 3n E2B IT is a multimodal, instruction-tuned model developed by Google DeepMind, designed to operate efficiently at an effective parameter size of 2B while leveraging a 6B architecture."},{"id":"google/gemma-3n-e4b-it:free","name":"Google: Gemma 3n 4B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3n","description":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.","context_window":8192,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","release_date":"2025-05-20","tagline":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets."},{"id":"google/gemma-3n-e4b-it","name":"Google: Gemma 3n 4B","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemma-3n","description":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.06,"output":0.12},"model_type":"chat","release_date":"2025-05-20","tagline":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"google/gemma-4-26b-a4b-it:free","name":"Google: Gemma 4 26B A4B (free)","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","release_date":"2026-04-03","context_window":262144,"max_output_tokens":32768,"parameters":26,"active_parameters":4,"model_type":"chat","tagline":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-08"},{"id":"google/gemma-4-26b-a4b-it","name":"Google: Gemma 4 26B A4B ","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","release_date":"2026-04-03","context_window":262144,"max_output_tokens":16384,"parameters":26,"active_parameters":4,"model_type":"chat","tagline":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.06,"output":0.33,"cached_input":0.04},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"google/gemma-4-31b-it:free","name":"Google: Gemma 4 31B (free)","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","release_date":"2026-04-02","context_window":262144,"max_output_tokens":8192,"parameters":31,"model_type":"chat","tagline":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"google/gemma-4-31b-it","name":"Google: Gemma 4 31B","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","release_date":"2026-04-02","context_window":262144,"max_output_tokens":262144,"parameters":31,"model_type":"chat","tagline":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"tool_call":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.12,"output":0.35,"cached_input":0.09},"last_updated":"2026-06-18","tools":["function_calling"]},{"id":"google/lyria-3-clip-preview","name":"Google: Lyria 3 Clip Preview","created_by":"google","source":"official","description":"30 second duration clips are priced at $0.04 per clip.","release_date":"2026-03-30","context_window":1048576,"max_output_tokens":65536,"model_type":"tts","tagline":"30 second duration clips are priced at $0.04 per clip.","capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text","audio"]},"last_updated":"2026-04-01","license":"proprietary","open_weight":false},{"id":"google/lyria-3-pro-preview","name":"Google: Lyria 3 Pro Preview","created_by":"google","source":"official","description":"Full-length songs are priced at $0.08 per song.","release_date":"2026-03-30","context_window":1048576,"max_output_tokens":65536,"model_type":"tts","tagline":"Full-length songs are priced at $0.08 per song.","capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text","audio"]},"last_updated":"2026-04-01","license":"proprietary","open_weight":false},{"id":"gryphe/mythomax-l2-13b","name":"MythoMax 13B","created_by":"gryphe","source":"official","last_updated":"2026-03-24","description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay.","context_window":4096,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.06,"output":0.06},"parameters":13,"release_date":"2023-07-02","tagline":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay."},{"id":"ibm-granite/granite-4.0-h-micro","name":"IBM: Granite 4.0 Micro","created_by":"ibm-granite","source":"official","last_updated":"2026-06-18","description":"Granite-4.0-H-Micro is a 3B parameter from the Granite 4 family of models.","context_window":131000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.017,"output":0.112},"release_date":"2025-10-20","tagline":"Granite-4.0-H-Micro is a 3B parameter from the Granite 4 family of models.","model_type":"chat","max_output_tokens":131000,"capabilities":{"structured_output":true,"json_mode":true}},{"id":"ibm-granite/granite-4.1-8b","name":"IBM: Granite 4.1 8B","created_by":"ibm-granite","source":"official","description":"Granite 4.1 8B is a dense, decoder-only 8-billion-parameter language model from IBM, part of the Granite 4.1 family.","release_date":"2026-04-30","context_window":131072,"max_output_tokens":131072,"parameters":8,"model_type":"chat","tagline":"Granite 4.1 8B is a dense, decoder-only 8-billion-parameter language model from IBM, part of the Granite 4.1 family.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.1,"cached_input":0.05},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"inception/mercury-2","name":"Inception: Mercury 2","created_by":"inception","source":"official","last_updated":"2026-03-24","description":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM).","context_window":128000,"max_output_tokens":50000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":0.75,"cached_input":0.025},"release_date":"2026-03-04","tagline":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM).","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"inception/mercury-coder","name":"Inception: Mercury Coder","created_by":"inception","source":"official","last_updated":"2026-04-03","description":"Mercury Coder is the first diffusion large language model (dLLM).","context_window":128000,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":0.75,"cached_input":0.025},"release_date":"2025-04-30","tagline":"Mercury Coder is the first diffusion large language model (dLLM).","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-15","license":"proprietary","open_weight":false},{"id":"inception/mercury","name":"Inception: Mercury","created_by":"inception","source":"official","last_updated":"2026-03-24","description":"Mercury is the first diffusion large language model (dLLM).","status":"deprecated","deprecation_date":"2026-04-15","context_window":128000,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":0.75,"cached_input":0.025},"release_date":"2025-06-26","tagline":"Mercury is the first diffusion large language model (dLLM).","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"inclusionai/ling-2.6-1t:free","name":"inclusionAI: Ling-2.6-1T (free)","created_by":"inclusionai","source":"official","description":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","status":"deprecated","release_date":"2026-04-23","deprecation_date":"2026-05-07","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"inclusionai/ling-2.6-1t","name":"inclusionAI: Ling-2.6-1T","created_by":"inclusionai","source":"official","description":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","release_date":"2026-04-23","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.075,"output":0.625,"cached_input":0.015},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"inclusionai/ling-2.6-flash:free","name":"inclusionAI: Ling-2.6-flash (free)","created_by":"inclusionai","source":"official","description":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","release_date":"2026-04-21","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-24","status":"deprecated","deprecation_date":"2026-04-29"},{"id":"inclusionai/ling-2.6-flash","name":"inclusionAI: Ling-2.6-flash","created_by":"inclusionai","source":"official","description":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","release_date":"2026-04-21","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.01,"output":0.03,"cached_input":0.002},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"inclusionai/ring-2.6-1t:free","name":"inclusionAI: Ring-2.6-1T (free)","created_by":"inclusionai","source":"official","description":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","release_date":"2026-05-08","context_window":262144,"max_output_tokens":65536,"tagline":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-10"},{"id":"inclusionai/ring-2.6-1t","name":"inclusionAI: Ring-2.6-1T","created_by":"inclusionai","source":"official","description":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","release_date":"2026-05-08","context_window":262144,"max_output_tokens":65536,"tagline":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.075,"output":0.625,"cached_input":0.015},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"inflection/inflection-3-pi","name":"Inflection: Inflection 3 Pi","created_by":"inflection","source":"official","last_updated":"2026-03-23","description":"Inflection 3 Pi powers Inflection's [Pi](https://pi.ai) chatbot, including backstory, emotional intelligence, productivity, and safety.","context_window":8000,"max_output_tokens":1024,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.5,"output":10},"release_date":"2024-10-11","tagline":"Inflection 3 Pi powers Inflection's [Pi](https://pi.ai) chatbot, including backstory, emotional intelligence, productivity, and safety."},{"id":"inflection/inflection-3-productivity","name":"Inflection: Inflection 3 Productivity","created_by":"inflection","source":"official","last_updated":"2026-03-23","description":"Inflection 3 Productivity is optimized for following instructions.","context_window":8000,"max_output_tokens":1024,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.5,"output":10},"release_date":"2024-10-11","tagline":"Inflection 3 Productivity is optimized for following instructions."},{"id":"kwaipilot/kat-coder-pro-v2","name":"Kwaipilot: KAT-Coder-Pro V2","created_by":"kwaipilot","source":"official","description":"KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and SaaS integration.","release_date":"2026-03-27","context_window":256000,"max_output_tokens":80000,"tagline":"KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and SaaS integration.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"kwaipilot/kat-coder-pro","name":"Kwaipilot: KAT-Coder-Pro V1","created_by":"kwaipilot","source":"official","last_updated":"2026-03-24","description":"KAT-Coder-Pro V1 is KwaiKAT's most advanced agentic coding model in the KAT-Coder series.","context_window":256000,"max_output_tokens":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.207,"output":0.828,"cached_input":0.041},"release_date":"2025-11-10","tagline":"KAT-Coder-Pro V1 is KwaiKAT's most advanced agentic coding model in the KAT-Coder series.","tools":["function_calling"]},{"id":"liquid/lfm-2-24b-a2b","name":"LiquidAI: LFM2-24B-A2B","created_by":"liquid","source":"official","last_updated":"2026-06-18","description":"LFM2-24B-A2B is the largest model in the LFM2 family of hybrid architectures designed for efficient on-device deployment.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.12},"parameters":24,"active_parameters":2,"release_date":"2026-02-25","tagline":"LFM2-24B-A2B is the largest model in the LFM2 family of hybrid architectures designed for efficient on-device deployment.","model_type":"chat","capabilities":{"structured_output":true,"json_mode":true}},{"id":"liquid/lfm-2.2-6b","name":"LiquidAI: LFM2-2.6B","created_by":"liquid","source":"official","last_updated":"2026-03-23","description":"LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.01,"output":0.02},"parameters":6,"release_date":"2025-10-20","tagline":"LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment.","model_type":"chat"},{"id":"liquid/lfm-2.5-1.2b-instruct:free","name":"LiquidAI: LFM2.5-1.2B-Instruct (free)","created_by":"liquid","source":"official","last_updated":"2026-06-18","description":"LFM2.5-1.2B-Instruct is a compact, high-performance instruction-tuned model built for fast on-device AI.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"parameters":1.2,"release_date":"2026-01-20","tagline":"LFM2.5-1.2B-Instruct is a compact, high-performance instruction-tuned model built for fast on-device AI.","model_type":"chat","capabilities":{"structured_output":true,"json_mode":true}},{"id":"liquid/lfm-2.5-1.2b-thinking:free","name":"LiquidAI: LFM2.5-1.2B-Thinking (free)","created_by":"liquid","source":"official","last_updated":"2026-06-18","description":"LFM2.5-1.2B-Thinking is a lightweight reasoning-focused model optimized for agentic tasks, data extraction, and RAG—while still running comfortably on edge devices.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true,"tool_call":true,"structured_output":true,"json_mode":true},"parameters":1.2,"release_date":"2026-01-20","tagline":"LFM2.5-1.2B-Thinking is a lightweight reasoning-focused model optimized for agentic tasks, data extraction, and RAG—while still running comfortably on edge devices.","model_type":"chat","tools":["function_calling"]},{"id":"liquid/lfm2-8b-a1b","name":"LiquidAI: LFM2-8B-A1B","created_by":"liquid","source":"official","last_updated":"2026-03-23","description":"LFM2-8B-A1B is an efficient on-device Mixture-of-Experts (MoE) model from Liquid AI’s LFM2 family, built for fast, high-quality inference on edge hardware.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.01,"output":0.02},"parameters":8,"active_parameters":1,"release_date":"2025-10-20","tagline":"LFM2-8B-A1B is an efficient on-device Mixture-of-Experts (MoE) model from Liquid AI’s LFM2 family, built for fast, high-quality inference on edge hardware.","model_type":"chat"},{"id":"mancer/weaver","name":"Mancer: Weaver (alpha)","created_by":"mancer","source":"official","last_updated":"2026-03-24","description":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory.","context_window":8000,"max_output_tokens":2000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.75,"output":1},"release_date":"2023-08-02","tagline":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory."},{"id":"meituan/longcat-flash-chat","name":"Meituan: LongCat Flash Chat","created_by":"meituan","source":"official","last_updated":"2026-03-24","description":"LongCat-Flash-Chat is a large-scale Mixture-of-Experts (MoE) model with 560B total parameters, of which 18.6B–31.3B (≈27B on average) are dynamically activated per input.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.8,"cached_input":0.2},"release_date":"2025-09-09","tagline":"LongCat-Flash-Chat is a large-scale Mixture-of-Experts (MoE) model with 560B total parameters, of which 18.6B–31.3B (≈27B on average) are dynamically activated per input.","tools":["function_calling"]},{"id":"meta-llama/llama-3-70b-instruct","name":"Meta: Llama 3 70B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","context_window":8192,"max_output_tokens":8000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.51,"output":0.74},"model_type":"chat","parameters":70,"release_date":"2024-04-18","tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","status":"deprecated","deprecation_date":"2026-06-19"},{"id":"meta-llama/llama-3-8b-instruct","name":"Meta: Llama 3 8B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","context_window":8192,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.14,"output":0.14},"model_type":"chat","parameters":8,"release_date":"2024-04-18","tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"meta-llama/llama-3.1-405b","name":"Meta: Llama 3.1 405B (base)","created_by":"meta-llama","source":"official","last_updated":"2026-03-23","family":"llama-3.1","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":4,"output":4},"model_type":"chat","parameters":405,"release_date":"2024-08-02","tagline":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors."},{"id":"meta-llama/llama-3.1-70b-instruct","name":"Meta: Llama 3.1 70B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-04-25","family":"llama-3.1","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":0.4},"model_type":"chat","parameters":70,"release_date":"2024-07-23","tagline":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","tools":["function_calling"],"max_output_tokens":16384},{"id":"meta-llama/llama-3.1-8b-instruct","name":"Meta: Llama 3.1 8B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3.1","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.02,"output":0.03},"model_type":"chat","parameters":8,"release_date":"2024-07-23","tagline":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","tools":["function_calling"]},{"id":"meta-llama/llama-3.2-11b-vision-instruct","name":"Meta: Llama 3.2 11B Vision Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3.2","description":"Llama 3.2 11B Vision is a multimodal model with 11 billion parameters, designed to handle tasks combining visual and textual data.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.345,"output":0.345},"model_type":"chat","parameters":11,"release_date":"2024-09-25","tagline":"Llama 3.2 11B Vision is a multimodal model with 11 billion parameters, designed to handle tasks combining visual and textual data."},{"id":"meta-llama/llama-3.2-1b-instruct","name":"Meta: Llama 3.2 1B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-05-23","family":"llama-3.2","description":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.027,"output":0.201},"model_type":"chat","parameters":1,"release_date":"2024-09-25","tagline":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis.","max_output_tokens":60000},{"id":"meta-llama/llama-3.2-3b-instruct:free","name":"Meta: Llama 3.2 3B Instruct (free)","created_by":"meta-llama","source":"official","last_updated":"2026-03-23","family":"llama-3.2","description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","parameters":3,"release_date":"2024-09-25","tagline":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization."},{"id":"meta-llama/llama-3.2-3b-instruct","name":"Meta: Llama 3.2 3B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-05-23","family":"llama-3.2","description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.051,"output":0.335},"model_type":"chat","parameters":3,"release_date":"2024-09-25","tagline":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.","max_output_tokens":80000},{"id":"meta-llama/llama-3.3-70b-instruct:free","name":"Meta: Llama 3.3 70B Instruct (free)","created_by":"meta-llama","source":"official","last_updated":"2026-05-17","family":"llama-3.3","description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"model_type":"chat","parameters":70,"release_date":"2024-12-06","tagline":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","tools":["function_calling"]},{"id":"meta-llama/llama-3.3-70b-instruct","name":"Meta: Llama 3.3 70B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-04-22","family":"llama-3.3","description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.32},"model_type":"chat","parameters":70,"release_date":"2024-12-06","tagline":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","tools":["function_calling"]},{"id":"meta-llama/llama-4-maverick","name":"Meta: Llama 4 Maverick","created_by":"meta-llama","source":"official","last_updated":"2026-04-07","family":"llama-4","description":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forw","context_window":1048576,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","release_date":"2025-04-05","tagline":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forw","tools":["function_calling"]},{"id":"meta-llama/llama-4-scout","name":"Meta: Llama 4 Scout","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-4","description":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","context_window":10000000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","release_date":"2025-04-05","tagline":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","tools":["function_calling"]},{"id":"meta-llama/llama-guard-3-8b","name":"Llama Guard 3 8B","created_by":"meta-llama","source":"official","last_updated":"2026-05-23","family":"llama-guard","description":"Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.484,"output":0.03},"model_type":"moderation","parameters":8,"release_date":"2025-02-12","tagline":"Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification.","max_output_tokens":131072},{"id":"meta-llama/llama-guard-4-12b:free","name":"Meta: Llama Guard 4 12B (free)","created_by":"meta-llama","source":"official","family":"llama-guard","description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","release_date":"2025-04-30","context_window":163840,"max_output_tokens":65000,"model_type":"chat","tagline":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","modalities":{"input":["image","text"],"output":["text"]},"last_updated":"2026-04-16"},{"id":"meta-llama/llama-guard-4-12b","name":"Meta: Llama Guard 4 12B","created_by":"meta-llama","source":"official","last_updated":"2026-04-25","family":"llama-guard","description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","context_window":163840,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.18,"output":0.18},"model_type":"moderation","parameters":12,"release_date":"2025-04-30","tagline":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","max_output_tokens":16384},{"id":"microsoft/phi-4-mini-instruct","name":"Microsoft: Phi 4 Mini Instruct","created_by":"microsoft","source":"official","description":"Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data.","release_date":"2025-10-17","context_window":131072,"max_output_tokens":128000,"model_type":"chat","tagline":"Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data.","capabilities":{"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.08,"output":0.35,"cached_input":0.08},"last_updated":"2026-05-17"},{"id":"microsoft/phi-4","name":"Microsoft: Phi 4","created_by":"microsoft","source":"official","last_updated":"2026-03-24","description":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed.","context_window":16384,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.065,"output":0.14},"release_date":"2025-01-10","tagline":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed.","model_type":"chat"},{"id":"microsoft/wizardlm-2-8x22b","name":"WizardLM-2 8x22B","created_by":"microsoft","source":"official","last_updated":"2026-05-17","description":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","context_window":65536,"max_output_tokens":8000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.62,"output":0.62},"release_date":"2024-04-16","tagline":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"minimax/minimax-01","name":"MiniMax: MiniMax-01","created_by":"minimax","source":"official","last_updated":"2026-03-23","description":"MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding.","context_window":1000192,"max_output_tokens":1000192,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":1.1},"model_type":"chat","family":"minimax","release_date":"2025-01-15","tagline":"MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding."},{"id":"minimax/minimax-m1","name":"MiniMax: MiniMax M1","created_by":"minimax","source":"official","last_updated":"2026-03-23","description":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference.","context_window":1000000,"max_output_tokens":40000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":0.4,"output":2.2},"model_type":"chat","family":"minimax","release_date":"2025-06-17","tagline":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference.","tools":["function_calling"]},{"id":"minimax/minimax-m2-her","name":"MiniMax: MiniMax M2-her","created_by":"minimax","source":"official","last_updated":"2026-03-23","description":"MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations.","context_window":65536,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2,"cached_input":0.03},"model_type":"chat","family":"minimax","release_date":"2026-01-23","tagline":"MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations."},{"id":"minimax/minimax-m2.1","name":"MiniMax: MiniMax M2.1","created_by":"minimax","source":"official","last_updated":"2026-05-17","description":"MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development.","context_window":204800,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.29,"output":0.95,"cached_input":0.03},"family":"minimax","model_type":"chat","release_date":"2025-12-23","tagline":"MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development.","tools":["function_calling"],"max_output_tokens":196608,"license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5:free","name":"MiniMax: MiniMax M2.5 (free)","created_by":"minimax","source":"official","last_updated":"2026-05-17","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","context_window":204800,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"family":"minimax","model_type":"chat","release_date":"2026-02-12","tagline":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","tools":["function_calling"]},{"id":"minimax/minimax-m2.5","name":"MiniMax: MiniMax M2.5","created_by":"minimax","source":"official","last_updated":"2026-06-10","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","context_window":204800,"max_output_tokens":196608,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.9,"cached_input":0.05},"family":"minimax","model_type":"chat","release_date":"2026-02-12","tagline":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","tools":["function_calling"],"license":"mit","open_weight":true},{"id":"minimax/minimax-m2.7","name":"MiniMax: MiniMax M2.7","created_by":"minimax","source":"official","last_updated":"2026-06-18","description":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.","context_window":204800,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":1,"cached_input":0.05},"family":"minimax","model_type":"chat","release_date":"2026-03-18","tagline":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2","name":"MiniMax: MiniMax M2","created_by":"minimax","source":"official","last_updated":"2026-05-17","description":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","context_window":204800,"max_output_tokens":196608,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":0.255,"output":1,"cached_input":0.03},"model_type":"chat","family":"minimax","release_date":"2025-10-23","tagline":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"minimax/minimax-m3","name":"MiniMax: MiniMax M3","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax-M3 is a multimodal foundation model from MiniMax.","release_date":"2026-05-31","context_window":1048576,"max_output_tokens":512000,"model_type":"chat","tagline":"MiniMax-M3 is a multimodal foundation model from MiniMax.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06},"tools":["function_calling"],"last_updated":"2026-06-02","license":"proprietary","open_weight":false},{"id":"mistralai/codestral-2508","name":"Mistral: Codestral 2508","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"codestral","description":"Mistral's cutting-edge language model for coding released end of July 2025.","context_window":256000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.3,"output":0.9,"cached_input":0.03},"model_type":"code","release_date":"2025-08-01","alias":"mistralai/codestral","tagline":"Mistral's cutting-edge language model for coding released end of July 2025.","tools":["function_calling"],"license":"mnpl","open_weight":true},{"id":"mistralai/codestral","name":"Mistral: Codestral 2508","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"codestral","description":"Mistral's cutting-edge language model for coding released end of July 2025.","release_date":"2025-08-01","context_window":256000,"model_type":"code","tagline":"Mistral's cutting-edge language model for coding released end of July 2025.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.9,"cached_input":0.03},"tools":["function_calling"],"snapshots":["mistralai/codestral-2508"],"license":"mnpl","open_weight":true},{"id":"mistralai/devstral-2512","name":"Mistral: Devstral 2 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"code","release_date":"2025-12-09","alias":"mistralai/devstral","tagline":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","tools":["function_calling"]},{"id":"mistralai/devstral-medium","name":"Mistral: Devstral Medium","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"chat","release_date":"2025-07-10","tagline":"Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"mistralai/devstral-small","name":"Mistral: Devstral Small 1.1","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"model_type":"chat","release_date":"2025-07-10","tagline":"Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/devstral","name":"Mistral: Devstral 2 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","release_date":"2025-12-09","context_window":262144,"model_type":"code","tagline":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"tools":["function_calling"],"snapshots":["mistralai/devstral-2512"]},{"id":"mistralai/ministral-14b-2512","name":"Mistral: Ministral 3 14B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.2,"cached_input":0.02},"model_type":"chat","parameters":14,"release_date":"2025-12-02","alias":"mistralai/ministral-14b","tagline":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-14b","name":"Mistral: Ministral 3 14B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","release_date":"2025-12-02","context_window":262144,"parameters":14,"model_type":"chat","tagline":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2,"cached_input":0.02},"tools":["function_calling"],"snapshots":["mistralai/ministral-14b-2512"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-3b-2512","name":"Mistral: Ministral 3 3B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.1,"cached_input":0.01},"model_type":"chat","parameters":3,"release_date":"2025-12-02","alias":"mistralai/ministral-3b","tagline":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-3b","name":"Mistral: Ministral 3 3B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","release_date":"2025-12-02","context_window":131072,"parameters":3,"model_type":"chat","tagline":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1,"cached_input":0.01},"tools":["function_calling"],"snapshots":["mistralai/ministral-3b-2512"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-8b-2512","name":"Mistral: Ministral 3 8B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15,"cached_input":0.015},"model_type":"chat","parameters":8,"release_date":"2025-12-02","alias":"mistralai/ministral-8b","tagline":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-8b","name":"Mistral: Ministral 3 8B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","release_date":"2025-12-02","context_window":262144,"parameters":8,"model_type":"chat","tagline":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15,"cached_input":0.015},"tools":["function_calling"],"snapshots":["mistralai/ministral-8b-2512"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-7b-instruct-v0.1","name":"Mistral: Mistral 7B Instruct v0.1","created_by":"mistral","source":"official","last_updated":"2026-05-23","description":"A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length.","context_window":4096,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.11,"output":0.19},"model_type":"chat","family":"mistral-7b","parameters":7,"release_date":"2023-09-28","tagline":"A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length.","status":"deprecated","deprecation_date":"2026-05-30","max_output_tokens":2824},{"id":"mistralai/mistral-large-2407","name":"Mistral Large 2407","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-large","description":"This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407).","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-11-19","alias":"mistralai/mistral-large","tagline":"This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407).","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-large-2411","name":"Mistral Large 2411","created_by":"mistral","source":"official","last_updated":"2026-04-07","family":"mistral-large","description":"Mistral Large 2 2411 is an update of [Mistral Large 2](/mistralai/mistral-large) released together with [Pixtral Large 2411](/mistralai/pixtral-large-2411) It provides a significant upgrade on the pre","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-11-19","alias":"mistralai/mistral-large","tagline":"Mistral Large 2 2411 is an update of [Mistral Large 2](/mistralai/mistral-large) released together with [Pixtral Large 2411](/mistralai/pixtral-large-2411)\n\nIt provides a significant upgrade on the pr","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-large-2512","name":"Mistral: Mistral Large 3 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-large","description":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.5,"output":1.5,"cached_input":0.05},"model_type":"chat","release_date":"2025-12-01","alias":"mistralai/mistral-large","tagline":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-large","name":"Mistral Large","created_by":"mistral","source":"official","last_updated":"2026-03-26","family":"mistral-large","description":"This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`).","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-02-26","tagline":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","tools":["function_calling"],"snapshots":["mistralai/mistral-large-2512","mistralai/mistral-large-2411","mistralai/mistral-large-2407"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-medium-3-5","name":"Mistral: Mistral Medium 3.5","created_by":"mistral","source":"official","family":"mistral-medium","description":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI.","release_date":"2026-04-30","context_window":262144,"model_type":"chat","tagline":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"tools":["function_calling"],"last_updated":"2026-05-06","license":"proprietary","open_weight":false},{"id":"mistralai/mistral-medium-3.1","name":"Mistral: Mistral Medium 3.1","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-medium","description":"Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"chat","release_date":"2025-08-13","tagline":"Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced opera","tools":["function_calling"]},{"id":"mistralai/mistral-medium-3","name":"Mistral: Mistral Medium 3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-medium","description":"Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"chat","release_date":"2025-05-07","tagline":"Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost.","tools":["function_calling"]},{"id":"mistralai/mistral-nemo","name":"Mistral: Mistral Nemo","created_by":"mistral","source":"official","last_updated":"2026-05-23","description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.02,"output":0.03},"model_type":"chat","family":"mistral-nemo","release_date":"2024-07-19","tagline":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-saba","name":"Mistral: Saba","created_by":"mistral","source":"official","last_updated":"2026-03-24","description":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.6,"cached_input":0.02},"model_type":"chat","family":"mistral-saba","release_date":"2025-02-17","tagline":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-small-24b-instruct-2501","name":"Mistral: Mistral Small 3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","context_window":32768,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.05,"output":0.08},"model_type":"chat","parameters":24,"release_date":"2025-01-30","alias":"mistralai/mistral-small-24b-instruct","tagline":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","tools":["function_calling"]},{"id":"mistralai/mistral-small-24b-instruct","name":"Mistral: Mistral Small 3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","release_date":"2025-01-30","context_window":32768,"max_output_tokens":16384,"parameters":24,"model_type":"chat","tagline":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.08},"tools":["function_calling"],"snapshots":["mistralai/mistral-small-24b-instruct-2501"]},{"id":"mistralai/mistral-small-2603","name":"Mistral: Mistral Small 4","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.015},"model_type":"chat","release_date":"2026-03-16","alias":"mistralai/mistral-small","tagline":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-small-3.1-24b-instruct:free","name":"Mistral: Mistral Small 3.1 24B (free)","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities.","status":"deprecated","deprecation_date":"2026-03-29","context_window":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"model_type":"chat","parameters":24,"release_date":"2025-03-17","tagline":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities.","tools":["function_calling"]},{"id":"mistralai/mistral-small-3.1-24b-instruct","name":"Mistral: Mistral Small 3.1 24B","created_by":"mistral","source":"official","last_updated":"2026-05-23","family":"mistral-small","description":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities.","context_window":128000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.351,"output":0.555,"cached_input":0.015},"model_type":"chat","parameters":24,"release_date":"2025-03-17","tagline":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities."},{"id":"mistralai/mistral-small-3.2-24b-instruct","name":"Mistral: Mistral Small 3.2 24B","created_by":"mistral","source":"official","last_updated":"2026-04-25","family":"mistral-small","description":"Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling.","context_window":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.075,"output":0.2},"model_type":"chat","parameters":24,"release_date":"2025-06-20","tagline":"Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling.","tools":["function_calling"],"max_output_tokens":16384},{"id":"mistralai/mistral-small-creative","name":"Mistral: Mistral Small Creative","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","description":"Mistral Small Creative is an experimental small model designed for creative writing, narrative generation, roleplay and character-driven dialogue, general-purpose instruction following, and conversati","status":"deprecated","deprecation_date":"2026-04-30","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"model_type":"chat","release_date":"2025-12-16","tagline":"Mistral Small Creative is an experimental small model designed for creative writing, narrative generation, roleplay and character-driven dialogue, general-purpose instruction following, and conversati","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-small","name":"Mistral: Mistral Small 4","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","release_date":"2026-03-16","context_window":262144,"model_type":"chat","tagline":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6,"cached_input":0.015},"tools":["function_calling"],"snapshots":["mistralai/mistral-small-2603"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mixtral-8x22b-instruct","name":"Mistral: Mixtral 8x22B Instruct","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mixtral","description":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b).","context_window":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-04-17","tagline":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b).","tools":["function_calling"]},{"id":"mistralai/mixtral-8x7b-instruct","name":"Mistral: Mixtral 8x7B Instruct","created_by":"mistral","source":"official","last_updated":"2026-05-07","family":"mixtral","description":"Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use.","context_window":32768,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.54,"output":0.54},"model_type":"chat","release_date":"2023-12-10","tagline":"Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"mistralai/pixtral-12b","name":"Mistral: Pixtral 12B","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"pixtral","description":"The first multi-modal, text+image-to-text model from Mistral AI.","context_window":32768,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","parameters":12,"release_date":"2024-09-10","tagline":"The first multi-modal, text+image-to-text model from Mistral AI.","license":"apache-2.0","open_weight":true,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true}},{"id":"mistralai/pixtral-large-2411","name":"Mistral: Pixtral Large 2411","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"pixtral","description":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-11-19","alias":"mistralai/pixtral-large","tagline":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","tools":["function_calling"],"license":"mnpl","open_weight":true},{"id":"mistralai/pixtral-large","name":"Mistral: Pixtral Large 2411","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"pixtral","description":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","release_date":"2024-11-19","context_window":131072,"model_type":"chat","tagline":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":6,"cached_input":0.2},"tools":["function_calling"],"snapshots":["mistralai/pixtral-large-2411"],"license":"mnpl","open_weight":true},{"id":"mistralai/voxtral-small-24b-2507","name":"Mistral: Voxtral Small 24B 2507","created_by":"mistral","source":"official","last_updated":"2026-03-24","description":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","context_window":32000,"modalities":{"input":["text","audio"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"model_type":"audio","family":"voxtral","parameters":24,"release_date":"2025-10-30","alias":"mistralai/voxtral-small-24b","tagline":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","tools":["function_calling"]},{"id":"mistralai/voxtral-small-24b","name":"Mistral: Voxtral Small 24B 2507","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"voxtral","description":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","release_date":"2025-10-30","context_window":32000,"parameters":24,"model_type":"audio","tagline":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text","audio"],"output":["text"]},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"tools":["function_calling"],"snapshots":["mistralai/voxtral-small-24b-2507"]},{"id":"moonshotai/kimi-k2-0905","name":"MoonshotAI: Kimi K2 0905","created_by":"moonshotai","source":"official","last_updated":"2026-05-14","description":"Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2).","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.6,"output":2.5,"cached_input":0.15},"family":"kimi-k2","model_type":"chat","release_date":"2025-09-04","tagline":"Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2).","tools":["function_calling"],"max_output_tokens":262144,"status":"deprecated","deprecation_date":"2026-05-14"},{"id":"moonshotai/kimi-k2-thinking","name":"MoonshotAI: Kimi K2 Thinking","created_by":"moonshotai","source":"official","last_updated":"2026-04-16","description":"Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":2.5,"cached_input":0.15},"family":"kimi-k2","model_type":"chat","release_date":"2025-11-06","tagline":"Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning.","tools":["function_calling"],"max_output_tokens":262144},{"id":"moonshotai/kimi-k2.5","name":"MoonshotAI: Kimi K2.5","created_by":"moonshotai","source":"official","last_updated":"2026-06-18","description":"Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.375,"output":2.025,"cached_input":0.09},"family":"kimi-k2.5","model_type":"chat","release_date":"2026-01-27","tagline":"Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm.","tools":["function_calling"]},{"id":"moonshotai/kimi-k2.6:free","name":"MoonshotAI: Kimi K2.6 (free)","created_by":"moonshotai","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","release_date":"2026-04-20","context_window":262144,"model_type":"chat","tagline":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"moonshotai/kimi-k2.6","name":"MoonshotAI: Kimi K2.6","created_by":"moonshotai","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","release_date":"2026-04-20","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.67,"output":3.5,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"moonshotai/kimi-k2.7-code","name":"MoonshotAI: Kimi K2.7 Code","created_by":"moonshotai","source":"official","family":"kimi-k2.7","description":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.","release_date":"2026-06-12","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.612,"output":3.069,"cached_input":0.13},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"moonshotai/kimi-k2","name":"MoonshotAI: Kimi K2 0711","created_by":"moonshotai","source":"official","last_updated":"2026-04-16","description":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.57,"output":2.3},"family":"kimi-k2","model_type":"chat","release_date":"2025-07-11","tagline":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass.","tools":["function_calling"],"max_output_tokens":32768},{"id":"morph/morph-v3-fast","name":"Morph: Morph V3 Fast","created_by":"morph","source":"official","last_updated":"2026-03-23","description":"Morph's fastest apply model for code edits.","context_window":81920,"max_output_tokens":38000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.2},"release_date":"2025-07-07","tagline":"Morph's fastest apply model for code edits."},{"id":"morph/morph-v3-large","name":"Morph: Morph V3 Large","created_by":"morph","source":"official","last_updated":"2026-06-20","description":"Morph's high-accuracy apply model for complex code edits.","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.9,"output":1.9},"release_date":"2025-07-07","tagline":"Morph's high-accuracy apply model for complex code edits.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"nex-agi/deepseek-v3.1-nex-n1","name":"Nex AGI: DeepSeek V3.1 Nex N1","created_by":"nex-agi","source":"official","last_updated":"2026-06-03","family":"deepseek","description":"DeepSeek V3.1 Nex-N1 is the flagship release of the Nex-N1 series — a post-trained model designed to highlight agent autonomy, tool use, and real-world productivity.","context_window":131072,"max_output_tokens":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.135,"output":0.5},"release_date":"2025-12-08","tagline":"DeepSeek V3.1 Nex-N1 is the flagship release of the Nex-N1 series — a post-trained model designed to highlight agent autonomy, tool use, and real-world productivity.","tools":["function_calling"],"model_type":"chat","status":"deprecated","deprecation_date":"2026-06-08"},{"id":"nex-agi/nex-n2-pro:free","name":"Nex AGI: Nex-N2-Pro (free)","created_by":"nex-agi","source":"official","description":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total.","release_date":"2026-06-08","context_window":262144,"max_output_tokens":262144,"tagline":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20","status":"deprecated","deprecation_date":"2026-06-22"},{"id":"nousresearch/hermes-2-pro-llama-3-8b","name":"NousResearch: Hermes 2 Pro - Llama-3 8B","created_by":"nousresearch","source":"official","last_updated":"2026-05-29","description":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced...","context_window":8192,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.14,"output":0.14},"model_type":"chat","parameters":8,"release_date":"2024-05-27","tagline":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mod","status":"deprecated","deprecation_date":"2026-06-05"},{"id":"nousresearch/hermes-3-llama-3.1-405b:free","name":"Nous: Hermes 3 405B Instruct (free)","created_by":"nousresearch","source":"official","last_updated":"2026-04-07","description":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","release_date":"2024-08-16","tagline":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren"},{"id":"nousresearch/hermes-3-llama-3.1-405b","name":"Nous: Hermes 3 405B Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-04-07","description":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":1,"output":1},"model_type":"chat","parameters":405,"release_date":"2024-08-16","tagline":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren"},{"id":"nousresearch/hermes-3-llama-3.1-70b","name":"Nous: Hermes 3 70B Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-06-10","description":"Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, rea","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.7,"output":0.7},"model_type":"chat","parameters":70,"release_date":"2024-08-18","tagline":"Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, rea","max_output_tokens":16384},{"id":"nousresearch/hermes-4-405b","name":"Nous: Hermes 4 405B","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","description":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1,"output":3},"model_type":"chat","parameters":405,"release_date":"2025-08-26","tagline":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research."},{"id":"nousresearch/hermes-4-70b","name":"Nous: Hermes 4 70B","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","description":"Hermes 4 70B is a hybrid reasoning model from Nous Research, built on Meta-Llama-3.1-70B.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.13,"output":0.4},"model_type":"chat","parameters":70,"release_date":"2025-08-26","tagline":"Hermes 4 70B is a hybrid reasoning model from Nous Research, built on Meta-Llama-3.1-70B."},{"id":"nvidia/llama-3.1-nemotron-70b-instruct","name":"NVIDIA: Llama 3.1 Nemotron 70B Instruct","created_by":"nvidia","source":"official","last_updated":"2026-05-07","family":"llama-3.1","description":"NVIDIA's Llama 3.1 Nemotron 70B is a language model designed for generating precise and useful responses.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.2,"output":1.2},"model_type":"chat","parameters":70,"release_date":"2024-10-15","tagline":"NVIDIA's Llama 3.1 Nemotron 70B is a language model designed for generating precise and useful responses.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"nvidia/llama-3.1-nemotron-ultra-253b-v1","name":"NVIDIA: Llama 3.1 Nemotron Ultra 253B v1","created_by":"nvidia","source":"official","last_updated":"2026-03-24","family":"llama-3.1","description":"Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) optimized for advanced reasoning, human-interactive chat, retrieval-augmented generation (RAG), and tool-calling tasks.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":1.8},"model_type":"chat","parameters":253,"release_date":"2025-04-08","tagline":"Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) optimized for advanced reasoning, human-interactive chat, retrieval-augmented generation (RAG), and tool-calling tasks."},{"id":"nvidia/llama-3.3-nemotron-super-49b-v1.5","name":"NVIDIA: Llama 3.3 Nemotron Super 49B V1.5","created_by":"nvidia","source":"official","last_updated":"2026-06-10","family":"llama-3.3","description":"Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.4,"output":0.4},"model_type":"chat","parameters":49,"release_date":"2025-10-10","tagline":"Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context.","tools":["function_calling"],"max_output_tokens":16384},{"id":"nvidia/nemotron-3-nano-30b-a3b:free","name":"NVIDIA: Nemotron 3 Nano 30B A3B (free)","created_by":"nvidia","source":"official","last_updated":"2026-03-23","description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","context_window":256000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-12-14","tagline":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","tools":["function_calling"]},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"NVIDIA: Nemotron 3 Nano 30B A3B","created_by":"nvidia","source":"official","last_updated":"2026-04-25","description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.05,"output":0.2},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-12-14","tagline":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","tools":["function_calling"],"max_output_tokens":228000},{"id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free","name":"NVIDIA: Nemotron 3 Nano Omni (free)","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems.","release_date":"2026-04-28","context_window":256000,"max_output_tokens":65536,"parameters":30,"active_parameters":3,"model_type":"reasoning","tagline":"NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text","audio","image","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"nvidia/nemotron-3-super-120b-a12b:free","name":"NVIDIA: Nemotron 3 Super (free)","created_by":"nvidia","source":"official","last_updated":"2026-05-17","description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","context_window":1000000,"max_output_tokens":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"model_type":"chat","parameters":120,"active_parameters":12,"release_date":"2026-03-11","tagline":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","tools":["function_calling"]},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"NVIDIA: Nemotron 3 Super","created_by":"nvidia","source":"official","last_updated":"2026-05-23","description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","context_window":1000000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.09,"output":0.45,"cached_input":0.1},"model_type":"chat","parameters":120,"active_parameters":12,"release_date":"2026-03-11","tagline":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","tools":["function_calling"],"max_output_tokens":16384},{"id":"nvidia/nemotron-3-ultra-550b-a55b:free","name":"NVIDIA: Nemotron 3 Ultra (free)","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","release_date":"2026-06-04","context_window":1000000,"max_output_tokens":65536,"parameters":550,"active_parameters":55,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"NVIDIA: Nemotron 3 Ultra","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","release_date":"2026-06-04","context_window":1000000,"max_output_tokens":16384,"parameters":550,"active_parameters":55,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":2.2,"cached_input":0.1},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"nvidia/nemotron-3.5-content-safety:free","name":"NVIDIA: Nemotron 3.5 Content Safety (free)","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B.","release_date":"2026-06-04","context_window":128000,"max_output_tokens":8192,"model_type":"chat","tagline":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B.","capabilities":{"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-nano-12b-v2-vl:free","name":"NVIDIA: Nemotron Nano 12B 2 VL (free)","created_by":"nvidia","source":"official","last_updated":"2026-03-23","description":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","context_window":128000,"max_output_tokens":128000,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"model_type":"chat","parameters":12,"release_date":"2025-10-28","tagline":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","tools":["function_calling"]},{"id":"nvidia/nemotron-nano-12b-v2-vl","name":"NVIDIA: Nemotron Nano 12B 2 VL","created_by":"nvidia","source":"official","last_updated":"2026-05-01","description":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","context_window":131072,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":12,"release_date":"2025-10-28","tagline":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","max_output_tokens":16384,"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"nvidia/nemotron-nano-9b-v2:free","name":"NVIDIA: Nemotron Nano 9B V2 (free)","created_by":"nvidia","source":"official","last_updated":"2026-03-24","description":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"model_type":"chat","parameters":9,"release_date":"2025-09-05","tagline":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","tools":["function_calling"]},{"id":"nvidia/nemotron-nano-9b-v2","name":"NVIDIA: Nemotron Nano 9B V2","created_by":"nvidia","source":"official","last_updated":"2026-06-10","description":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.04,"output":0.16},"model_type":"chat","parameters":9,"release_date":"2025-09-05","tagline":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","tools":["function_calling"],"max_output_tokens":16384,"status":"deprecated","deprecation_date":"2026-06-11"},{"id":"openai/gpt-3.5-turbo-0613","name":"OpenAI: GPT-3.5 Turbo (older v0613)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"GPT-3.5 Turbo is OpenAI's fastest model.","context_window":4095,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":true,"streaming":false,"vision":false,"reasoning":false,"batch":true},"pricing":{"input":1,"output":2},"model_type":"chat","release_date":"2024-01-25","tagline":"GPT-3.5 Turbo is OpenAI's fastest model.","tools":["function_calling"],"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-3.5-turbo-16k","name":"OpenAI: GPT-3.5 Turbo 16k","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost.","context_window":16385,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":3,"output":4},"model_type":"chat","release_date":"2023-08-28","tagline":"This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost.","tools":["function_calling"]},{"id":"openai/gpt-3.5-turbo-instruct","name":"OpenAI: GPT-3.5 Turbo Instruct","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations.","context_window":4095,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"fine_tuning":true,"streaming":false,"vision":false,"tool_call":false,"reasoning":false},"pricing":{"input":1.5,"output":2},"model_type":"chat","release_date":"2023-09-28","tagline":"This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations.","knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-3.5-turbo","name":"OpenAI: GPT-3.5 Turbo","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"GPT-3.5 Turbo is OpenAI's fastest model.","context_window":16385,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":true,"streaming":false,"vision":false,"reasoning":false,"batch":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","release_date":"2023-05-28","tagline":"GPT-3.5 Turbo is OpenAI's fastest model.","tools":["function_calling"],"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-4-0314","name":"OpenAI: GPT-4 (older v0314)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4","description":"GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14.","context_window":8191,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":30,"output":60},"model_type":"chat","release_date":"2023-05-28","tagline":"GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14.","tools":["function_calling"]},{"id":"openai/gpt-4-1106-preview","name":"OpenAI: GPT-4 Turbo (older v1106)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4","description":"The latest GPT-4 Turbo model with vision capabilities.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":30},"model_type":"chat","release_date":"2023-11-06","tagline":"The latest GPT-4 Turbo model with vision capabilities.","tools":["function_calling"]},{"id":"openai/gpt-4-turbo-preview","name":"OpenAI: GPT-4 Turbo Preview","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4-turbo","description":"The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":30},"model_type":"chat","release_date":"2024-01-25","tagline":"The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"openai/gpt-4-turbo","name":"OpenAI: GPT-4 Turbo","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4-turbo","description":"The latest GPT-4 Turbo model with vision capabilities.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":10,"output":30},"model_type":"chat","release_date":"2024-04-09","tagline":"The latest GPT-4 Turbo model with vision capabilities.","tools":["function_calling"],"knowledge_cutoff":"2023-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-mini","name":"OpenAI: GPT-4.1 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4.1","description":"GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost.","context_window":1047576,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.4,"output":1.6,"cached_input":0.1},"model_type":"chat","release_date":"2025-04-14","tagline":"GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-nano","name":"OpenAI: GPT-4.1 Nano","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4.1","description":"For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series.","context_window":1047576,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.025},"model_type":"chat","release_date":"2025-04-14","tagline":"For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1","name":"OpenAI: GPT-4.1","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4.1","description":"GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning.","context_window":1047576,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":2,"output":8,"cached_input":0.5},"model_type":"chat","release_date":"2025-04-14","tagline":"GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4","name":"OpenAI: GPT-4","created_by":"openai","source":"official","last_updated":"2026-04-07","family":"gpt-4","description":"OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and adva","context_window":8191,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"vision":false,"reasoning":false,"batch":true},"pricing":{"input":30,"output":60},"model_type":"chat","release_date":"2023-05-28","tagline":"OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and adva","tools":["function_calling"],"knowledge_cutoff":"2023-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-2024-05-13","name":"OpenAI: GPT-4o (2024-05-13)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":5,"output":15,"cached_input":1.25},"model_type":"chat","release_date":"2024-05-13","alias":"openai/gpt-4o","tagline":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-2024-08-06","name":"OpenAI: GPT-4o (2024-08-06)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2024-08-06","alias":"openai/gpt-4o","tagline":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-2024-11-20","name":"OpenAI: GPT-4o (2024-11-20)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2024-11-20","alias":"openai/gpt-4o","tagline":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-audio-preview","name":"OpenAI: GPT-4o Audio","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"The gpt-4o-audio-preview model adds support for audio inputs as prompts.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["audio","text"],"output":["text","audio"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":2.5,"output":10},"model_type":"chat","release_date":"2025-08-15","tagline":"The gpt-4o-audio-preview model adds support for audio inputs as prompts.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o:extended","name":"OpenAI: GPT-4o (extended)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":6,"output":18},"model_type":"chat","release_date":"2024-05-13","tagline":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","tools":["function_calling"]},{"id":"openai/gpt-4o-mini-2024-07-18","name":"OpenAI: GPT-4o-mini (2024-07-18)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.075},"model_type":"chat","release_date":"2024-07-18","alias":"openai/gpt-4o-mini","tagline":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini-search-preview","name":"OpenAI: GPT-4o-mini Search Preview","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o mini Search Preview is a specialized model for web search in Chat Completions.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"tool_call":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","release_date":"2025-03-12","tagline":"GPT-4o mini Search Preview is a specialized model for web search in Chat Completions.","knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini","name":"OpenAI: GPT-4o-mini","created_by":"openai","source":"official","last_updated":"2026-06-03","family":"gpt-4o","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.075},"model_type":"chat","release_date":"2024-07-18","tagline":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","tools":["function_calling"],"snapshots":["openai/gpt-4o-mini-2024-07-18"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-search-preview","name":"OpenAI: GPT-4o Search Preview","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o Search Previewis a specialized model for web search in Chat Completions.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"tool_call":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":2.5,"output":10},"model_type":"chat","release_date":"2025-03-12","tagline":"GPT-4o Search Previewis a specialized model for web search in Chat Completions.","knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o","name":"OpenAI: GPT-4o","created_by":"openai","source":"official","last_updated":"2026-06-03","family":"gpt-4o","description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2024-05-13","tagline":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability.","tools":["function_calling"],"snapshots":["openai/gpt-4o-2024-11-20","openai/gpt-4o-2024-08-06","openai/gpt-4o-2024-05-13"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-chat","name":"OpenAI: GPT-5 Chat","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications."},{"id":"openai/gpt-5-codex","name":"OpenAI: GPT-5 Codex","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5-Codex is a specialized version of GPT-5 optimized for software engineering and coding workflows.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-09-23","tagline":"GPT-5-Codex is a specialized version of GPT-5 optimized for software engineering and coding workflows.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-image-mini","name":"OpenAI: GPT-5 Image Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2.5,"output":2,"cached_input":0.25},"model_type":"chat","release_date":"2025-10-16","tagline":"GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation.","tools":["function_calling"]},{"id":"openai/gpt-5-image","name":"OpenAI: GPT-5 Image","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":10,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2025-10-14","tagline":"[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities.","tools":["function_calling"]},{"id":"openai/gpt-5-mini","name":"OpenAI: GPT-5 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-nano","name":"OpenAI: GPT-5 Nano","created_by":"openai","source":"official","last_updated":"2026-04-12","family":"gpt-5","description":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.05,"output":0.4,"cached_input":0.01},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-pro","name":"OpenAI: GPT-5 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":120},"model_type":"chat","release_date":"2025-10-06","tagline":"GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-chat","name":"OpenAI: GPT-5.1 Chat","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.1","description":"GPT-5.1 Chat (AKA Instant is the fast, lightweight member of the 5.1 family, optimized for low-latency chat while retaining strong general intelligence.","context_window":128000,"max_output_tokens":32000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.25,"output":10,"cached_input":0.13},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1 Chat (AKA Instant is the fast, lightweight member of the 5.1 family, optimized for low-latency chat while retaining strong general intelligence.","tools":["function_calling"]},{"id":"openai/gpt-5.1-codex-max","name":"OpenAI: GPT-5.1-Codex-Max","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.1","description":"GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-12-04","tagline":"GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex-mini","name":"OpenAI: GPT-5.1-Codex-Mini","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.1","description":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","context_window":400000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex","name":"OpenAI: GPT-5.1-Codex","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.1","description":"GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.13},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1","name":"OpenAI: GPT-5.1","created_by":"openai","source":"official","last_updated":"2026-04-12","family":"gpt-5.1","description":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10,"cached_input":0.13},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2-chat","name":"OpenAI: GPT-5.2 Chat","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.2","description":"GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2025-12-10","tagline":"GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-08-10"},{"id":"openai/gpt-5.2-codex","name":"OpenAI: GPT-5.2-Codex","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.2","description":"GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2026-01-14","tagline":"GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2-pro","name":"OpenAI: GPT-5.2 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.2","description":"GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":21,"output":168},"model_type":"chat","release_date":"2025-12-10","tagline":"GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2","name":"OpenAI: GPT-5.2","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.2","description":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2025-12-10","tagline":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.3-chat","name":"OpenAI: GPT-5.3 Chat","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.3","description":"GPT-5.3 Chat is an update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2026-03-03","tagline":"GPT-5.3 Chat is an update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful.","tools":["function_calling"]},{"id":"openai/gpt-5.3-codex","name":"OpenAI: GPT-5.3-Codex","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.3","description":"GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2026-02-24","tagline":"GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilitie","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-image-2","name":"OpenAI: GPT-5.4 Image 2","created_by":"openai","source":"official","family":"gpt-5.4","description":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2.","release_date":"2026-04-21","context_window":272000,"max_output_tokens":128000,"model_type":"chat","tagline":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text"],"output":["image","text"]},"pricing":{"input":8,"output":15,"cached_input":2},"last_updated":"2026-04-22"},{"id":"openai/gpt-5.4-mini","name":"OpenAI: GPT-5.4 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5,"cached_input":0.075},"model_type":"chat","release_date":"2026-03-17","tagline":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-nano","name":"OpenAI: GPT-5.4 Nano","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25,"cached_input":0.02},"model_type":"chat","release_date":"2026-03-17","tagline":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-pro","name":"OpenAI: GPT-5.4 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks.","context_window":1050000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":30,"output":180},"model_type":"chat","release_date":"2026-03-05","tagline":"GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4","name":"OpenAI: GPT-5.4","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system.","context_window":1050000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15,"cached_input":0.25},"model_type":"chat","release_date":"2026-03-05","tagline":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5-pro","name":"OpenAI: GPT-5.5 Pro","created_by":"openai","source":"official","family":"gpt-5.5","description":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads.","release_date":"2026-04-24","context_window":1050000,"max_output_tokens":128000,"model_type":"chat","tagline":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":false,"vision":true,"fine_tuning":false,"batch":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":30,"output":180},"tools":["function_calling"],"last_updated":"2026-04-25","knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5","name":"OpenAI: GPT-5.5","created_by":"openai","source":"official","family":"gpt-5.5","description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks.","release_date":"2026-04-24","context_window":1050000,"max_output_tokens":128000,"model_type":"chat","tagline":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-04-25","knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5","name":"OpenAI: GPT-5","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-audio-mini","name":"OpenAI: GPT Audio Mini","created_by":"openai","source":"official","last_updated":"2026-04-01","family":"gpt-audio","description":"A cost-efficient version of GPT Audio.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","audio"],"output":["text","audio"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true,"streaming":false,"vision":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.6,"output":2.4},"release_date":"2026-01-19","tagline":"A cost-efficient version of GPT Audio.","model_type":"chat","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-audio","name":"OpenAI: GPT Audio","created_by":"openai","source":"official","last_updated":"2026-04-01","family":"gpt-audio","description":"The gpt-audio model is OpenAI's first generally available audio model.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","audio"],"output":["text","audio"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true,"streaming":true,"vision":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":2.5,"output":10},"release_date":"2026-01-19","tagline":"The gpt-audio model is OpenAI's first generally available audio model.","model_type":"chat","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-chat-latest","name":"OpenAI: GPT Chat Latest","created_by":"openai","source":"official","description":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT.","release_date":"2026-05-05","context_window":400000,"max_output_tokens":128000,"tagline":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-05-06"},{"id":"openai/gpt-oss-120b:free","name":"OpenAI: gpt-oss-120b (free)","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"model_type":"chat","release_date":"2025-08-05","tagline":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","tools":["function_calling"]},{"id":"openai/gpt-oss-120b","name":"OpenAI: gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-05-23","family":"gpt-oss","description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":false,"vision":false,"batch":true},"pricing":{"input":0.039,"output":0.18},"model_type":"chat","parameters":120,"release_date":"2025-08-05","tagline":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","tools":["function_calling"],"max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b:free","name":"OpenAI: gpt-oss-20b (free)","created_by":"openai","source":"official","last_updated":"2026-06-18","family":"gpt-oss","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"model_type":"chat","release_date":"2025-08-05","tagline":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","tools":["function_calling"]},{"id":"openai/gpt-oss-20b","name":"OpenAI: gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-06-02","family":"gpt-oss","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":false,"vision":false,"batch":true},"pricing":{"input":0.029,"output":0.14,"cached_input":0.015},"model_type":"chat","parameters":20,"release_date":"2025-08-05","tagline":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-safeguard-20b","name":"OpenAI: gpt-oss-safeguard-20b","created_by":"openai","source":"official","last_updated":"2026-06-20","family":"gpt-oss","description":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b.","context_window":131072,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.075,"output":0.3,"cached_input":0.038},"model_type":"moderation","parameters":20,"release_date":"2025-10-29","tagline":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b.","tools":["function_calling"]},{"id":"openai/o1-pro","name":"OpenAI: o1-pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o1","description":"The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"vision":true,"tool_call":true,"streaming":false,"fine_tuning":false,"batch":true},"pricing":{"input":150,"output":600},"model_type":"reasoning","release_date":"2025-03-19","tagline":"The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o1","name":"OpenAI: o1","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o1","description":"The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":60,"cached_input":7.5},"model_type":"reasoning","release_date":"2024-12-17","tagline":"The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-deep-research","name":"OpenAI: o3 Deep Research","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":10,"output":40,"cached_input":2.5},"model_type":"chat","release_date":"2025-10-10","tagline":"o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3-mini-high","name":"OpenAI: o3 Mini High","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.55},"model_type":"reasoning","release_date":"2025-02-12","tagline":"OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high.","tools":["function_calling"]},{"id":"openai/o3-mini","name":"OpenAI: o3 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":false,"fine_tuning":false,"batch":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.55},"model_type":"reasoning","release_date":"2025-01-31","tagline":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-pro","name":"OpenAI: o3 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"vision":true,"streaming":false,"fine_tuning":false,"batch":true},"pricing":{"input":20,"output":80},"model_type":"reasoning","release_date":"2025-06-10","tagline":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3","name":"OpenAI: o3","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"o3 is a well-rounded and powerful model across domains.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":2,"output":8,"cached_input":0.5},"model_type":"reasoning","release_date":"2025-04-16","tagline":"o3 is a well-rounded and powerful model across domains.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o4-mini-deep-research","name":"OpenAI: o4 Mini Deep Research","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o4","description":"o4-mini-deep-research is OpenAI's faster, more affordable deep research model—ideal for tackling complex, multi-step research tasks.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":2,"output":8,"cached_input":0.5},"model_type":"chat","release_date":"2025-10-10","tagline":"o4-mini-deep-research is OpenAI's faster, more affordable deep research model—ideal for tackling complex, multi-step research tasks.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o4-mini-high","name":"OpenAI: o4 Mini High","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o4","description":"OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.275},"model_type":"reasoning","release_date":"2025-04-16","tagline":"OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high.","tools":["function_calling"]},{"id":"openai/o4-mini","name":"OpenAI: o4 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o4","description":"OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"batch":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.275},"model_type":"reasoning","release_date":"2025-04-16","tagline":"OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openrouter/auto","name":"Auto Router","created_by":"openrouter","source":"official","last_updated":"2026-05-10","description":"Your prompt will be processed by a meta-model and routed to one of dozens of models (see below), optimizing for the best possible output.","context_window":2000000,"modalities":{"input":["text","image","audio","video"],"output":["text","image"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"release_date":"2023-11-08","tagline":"Your prompt will be processed by a meta-model and routed to one of dozens of models (see below), optimizing for the best possible output.","tools":["function_calling"]},{"id":"openrouter/bodybuilder","name":"Body Builder (beta)","created_by":"openrouter","source":"official","last_updated":"2026-03-23","description":"Transform your natural language requests into structured OpenRouter API request objects.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"release_date":"2025-12-05","tagline":"Transform your natural language requests into structured OpenRouter API request objects."},{"id":"openrouter/elephant-alpha","name":"Elephant","created_by":"openrouter","source":"official","description":"Elephant Alpha is a 100B-parameter text model focused on intelligence efficiency, delivering strong performance while minimizing token usage.","release_date":"2026-04-13","context_window":262144,"max_output_tokens":32768,"tagline":"Elephant Alpha is a 100B-parameter text model focused on intelligence efficiency, delivering strong reasoning performance while minimizing token usage.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-16"},{"id":"openrouter/free","name":"Free Models Router","created_by":"openrouter","source":"official","last_updated":"2026-03-24","description":"The simplest way to get free inference.","context_window":200000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"release_date":"2026-02-01","tagline":"The simplest way to get free inference.","tools":["function_calling"]},{"id":"openrouter/fusion","name":"OpenRouter: Fusion","created_by":"openrouter","source":"official","description":"Fusion turns your prompt into a small multi-model deliberation.","release_date":"2026-06-13","context_window":1000000,"tagline":"Fusion turns your prompt into a small multi-model deliberation.","modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"openrouter/owl-alpha","name":"Owl Alpha","created_by":"openrouter","source":"official","description":"Owl Alpha is a high-performance foundation model designed for agentic workloads.","release_date":"2026-04-28","context_window":1048756,"max_output_tokens":262144,"tagline":"Owl Alpha is a high-performance foundation model designed for agentic workloads.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"openrouter/pareto-code","name":"Pareto Code Router","created_by":"openrouter","source":"official","description":"The Pareto Router maintains a tiered shortlist of strong coding models, ranked by [Artificial Analysis](https://artificialanalysis.ai/) coding percentiles.","release_date":"2026-04-21","context_window":2000000,"tagline":"The Pareto Router is a way to have OpenRouter always pick a strong coding model for your needs without committing to a specific one.","modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"perceptron/perceptron-mk1","name":"Perceptron: Perceptron Mk1","created_by":"perceptron","source":"official","description":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces","release_date":"2026-05-12","context_window":32768,"max_output_tokens":8192,"tagline":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.15,"output":1.5},"last_updated":"2026-05-14"},{"id":"perplexity/sonar-deep-research","name":"Perplexity: Sonar Deep Research","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":2,"output":8},"family":"sonar","model_type":"chat","release_date":"2025-03-07","tagline":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics."},{"id":"perplexity/sonar-pro-search","name":"Perplexity: Sonar Pro Search","created_by":"perplexity","source":"official","last_updated":"2026-03-24","description":"Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system.","context_window":200000,"max_output_tokens":8000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15},"family":"sonar","model_type":"chat","release_date":"2025-10-30","tagline":"Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system."},{"id":"perplexity/sonar-pro","name":"Perplexity: Sonar Pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Note: Sonar Pro pricing includes Perplexity search pricing.","context_window":200000,"max_output_tokens":8000,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":3,"output":15},"family":"sonar","model_type":"chat","release_date":"2025-03-07","tagline":"Note: Sonar Pro pricing includes Perplexity search pricing."},{"id":"perplexity/sonar-reasoning-pro","name":"Perplexity: Sonar Reasoning Pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Note: Sonar Pro pricing includes Perplexity search pricing.","context_window":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":2,"output":8},"model_type":"reasoning","family":"sonar","release_date":"2025-03-07","tagline":"Note: Sonar Pro pricing includes Perplexity search pricing."},{"id":"perplexity/sonar","name":"Perplexity: Sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources.","context_window":127072,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1,"output":1},"family":"sonar","model_type":"chat","release_date":"2025-01-27","tagline":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources."},{"id":"poolside/laguna-m.1:free","name":"Poolside: Laguna M.1 (free)","created_by":"poolside","source":"official","description":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai/), optimized for complex software engineering tasks.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai), optimized for complex software engineering tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"poolside/laguna-m.1","name":"Poolside: Laguna M.1","created_by":"poolside","source":"official","description":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai/), optimized for complex software engineering tasks.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai/), optimized for complex software engineering tasks.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.2,"output":0.4,"cached_input":0.1},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"poolside/laguna-xs.2:free","name":"Poolside: Laguna XS.2 (free)","created_by":"poolside","source":"official","description":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai/), their efficient coding agent series.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai), their efficient coding agent series.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"poolside/laguna-xs.2","name":"Poolside: Laguna XS.2","created_by":"poolside","source":"official","description":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai/), their efficient coding agent series.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai/), their efficient coding agent series.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.2,"cached_input":0.05},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"prime-intellect/intellect-3","name":"Prime Intellect: INTELLECT-3","created_by":"prime-intellect","source":"official","last_updated":"2026-03-24","description":"INTELLECT-3 is a 106B-parameter Mixture-of-Experts model (12B active) post-trained from GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":1.1},"release_date":"2025-11-27","tagline":"INTELLECT-3 is a 106B-parameter Mixture-of-Experts model (12B active) post-trained from GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).","tools":["function_calling"]},{"id":"qwen/qwen-2.5-72b-instruct","name":"Qwen2.5 72B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen","description":"Qwen2.5 72B is the latest series of Qwen large language models.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.36,"output":0.4},"model_type":"chat","parameters":72,"release_date":"2024-09-19","tagline":"Qwen2.5 72B is the latest series of Qwen large language models.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen-2.5-7b-instruct","name":"Qwen: Qwen2.5 7B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen","description":"Qwen2.5 7B is the latest series of Qwen large language models.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.04,"output":0.1},"model_type":"chat","parameters":7,"release_date":"2024-10-16","tagline":"Qwen2.5 7B is the latest series of Qwen large language models.","tools":["function_calling"],"max_output_tokens":32768,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen-2.5-coder-32b-instruct","name":"Qwen2.5 Coder 32B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-23","family":"qwen","description":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen).","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.66,"output":1},"model_type":"code","parameters":32,"release_date":"2024-11-11","tagline":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen).","max_output_tokens":32768},{"id":"qwen/qwen-2.5-vl-7b-instruct","name":"Qwen: Qwen2.5-VL 7B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen","description":"Qwen2.5 VL 7B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\n- SoTA understanding of images of various resolution & ratio: Qwen2.5-VL achieves state-of-the-art performanc","context_window":32768,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"model_type":"chat","parameters":7,"release_date":"2024-08-28","tagline":"Qwen2.5 VL 7B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\n- SoTA understanding of images of various resolution & ratio: Qwen2.5-VL achieves state-of-the-art performanc","license":"apache-2.0","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true}},{"id":"qwen/qwen-max","name":"Qwen: Qwen-Max ","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen-Max, based on Qwen2.5, provides the best inference performance among [Qwen models](/qwen), especially for complex multi-step tasks.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true},"pricing":{"input":1.04,"output":4.16,"cached_input":0.208},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen-Max, based on Qwen2.5, provides the best inference performance among [Qwen models](/qwen), especially for complex multi-step tasks.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"qwen/qwen-plus-2025-07-28:thinking","name":"Qwen: Qwen Plus 0728 (thinking)","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","context_window":1000000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.26,"output":0.78},"model_type":"chat","release_date":"2025-09-08","tagline":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","tools":["function_calling"]},{"id":"qwen/qwen-plus-2025-07-28","name":"Qwen: Qwen Plus 0728","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","context_window":1000000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.26,"output":0.78,"cached_input":0.052},"model_type":"chat","release_date":"2025-09-08","alias":"qwen/qwen-plus","tagline":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","tools":["function_calling"]},{"id":"qwen/qwen-plus","name":"Qwen: Qwen-Plus","created_by":"qwen","source":"official","last_updated":"2026-03-26","family":"qwen","description":"Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.","context_window":1000000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true},"pricing":{"input":0.26,"output":0.78,"cached_input":0.052},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","tools":["function_calling"],"snapshots":["qwen/qwen-plus-2025-07-28"],"license":"proprietary","open_weight":false},{"id":"qwen/qwen-turbo","name":"Qwen: Qwen-Turbo","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen-Turbo, based on Qwen2.5, is a 1M context model that provides fast speed and low cost, suitable for simple tasks.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true},"pricing":{"input":0.033,"output":0.13,"cached_input":0.007},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen-Turbo, based on Qwen2.5, is a 1M context model that provides fast speed and low cost, suitable for simple tasks.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"qwen/qwen-vl-max","name":"Qwen: Qwen VL Max","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen VL Max is a visual understanding model with 7500 tokens context length.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.52,"output":2.08},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen VL Max is a visual understanding model with 7500 tokens context length.","tools":["function_calling"]},{"id":"qwen/qwen-vl-plus","name":"Qwen: Qwen VL Plus","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen's Enhanced Large Visual Language Model.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.137,"output":0.41,"cached_input":0.027},"model_type":"chat","release_date":"2025-02-05","tagline":"Qwen's Enhanced Large Visual Language Model."},{"id":"qwen/qwen2.5-coder-7b-instruct","name":"Qwen: Qwen2.5 Coder 7B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen2.5","description":"Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.03,"output":0.09},"model_type":"code","parameters":7,"release_date":"2025-04-15","tagline":"Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing."},{"id":"qwen/qwen2.5-vl-32b-instruct","name":"Qwen: Qwen2.5 VL 32B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen2.5","description":"Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities.","context_window":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"tool_call":true,"batch":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":32,"release_date":"2025-03-24","tagline":"Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities.","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen2.5-vl-72b-instruct","name":"Qwen: Qwen2.5 VL 72B Instruct","created_by":"qwen","source":"official","last_updated":"2026-06-18","family":"qwen2.5","description":"Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects.","context_window":131072,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"tool_call":true,"batch":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.8,"output":1,"cached_input":0.4},"model_type":"chat","parameters":72,"release_date":"2025-02-01","tagline":"Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects.","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-14b","name":"Qwen: Qwen3 14B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue.","context_window":131702,"max_output_tokens":40960,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.24},"model_type":"chat","parameters":14,"release_date":"2025-04-28","tagline":"Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-235b-a22b-2507","name":"Qwen: Qwen3 235B A22B Instruct 2507","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.09,"output":0.1},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-07-21","max_output_tokens":16384,"alias":"qwen/qwen3-235b-a22b","tagline":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","tools":["function_calling"]},{"id":"qwen/qwen3-235b-a22b-thinking-2507","name":"Qwen: Qwen3 235B A22B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-06-02","family":"qwen3","description":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.1,"cached_input":0.1},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-07-25","alias":"qwen/qwen3-235b-a22b-thinking","tagline":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","tools":["function_calling"],"max_output_tokens":262144,"status":"deprecated","deprecation_date":"2026-04-22","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-235b-a22b-thinking","name":"Qwen: Qwen3 235B A22B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-06-02","family":"qwen3","description":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","release_date":"2025-07-25","context_window":262144,"parameters":235,"active_parameters":22,"model_type":"chat","tagline":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.1,"cached_input":0.1},"tools":["function_calling"],"snapshots":["qwen/qwen3-235b-a22b-thinking-2507"],"max_output_tokens":262144,"status":"deprecated","deprecation_date":"2026-04-22","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-235b-a22b","name":"Qwen: Qwen3 235B A22B","created_by":"qwen","source":"official","last_updated":"2026-03-26","family":"qwen3","description":"Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.455,"output":1.82},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-04-28","tagline":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","tools":["function_calling"],"snapshots":["qwen/qwen3-235b-a22b-2507"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-instruct-2507","name":"Qwen: Qwen3 30B A3B Instruct 2507","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","context_window":131072,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.048,"output":0.193},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-07-29","alias":"qwen/qwen3-30b-a3b-instruct","tagline":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-instruct","name":"Qwen: Qwen3 30B A3B Instruct 2507","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","release_date":"2025-07-29","context_window":131072,"max_output_tokens":32000,"parameters":30,"active_parameters":3,"model_type":"chat","tagline":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.048,"output":0.193},"tools":["function_calling"],"snapshots":["qwen/qwen3-30b-a3b-instruct-2507"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-thinking-2507","name":"Qwen: Qwen3 30B A3B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.08,"output":0.4,"cached_input":0.08},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-08-28","alias":"qwen/qwen3-30b-a3b-thinking","tagline":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-thinking","name":"Qwen: Qwen3 30B A3B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","release_date":"2025-08-28","context_window":131072,"max_output_tokens":131072,"parameters":30,"active_parameters":3,"model_type":"chat","tagline":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.08,"output":0.4,"cached_input":0.08},"tools":["function_calling"],"snapshots":["qwen/qwen3-30b-a3b-thinking-2507"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b","name":"Qwen: Qwen3 30B A3B","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.12,"output":0.5},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-04-28","tagline":"Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tas","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-06-05","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-32b","name":"Qwen: Qwen3 32B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.08,"output":0.28,"cached_input":0.04},"model_type":"chat","parameters":32,"release_date":"2025-04-28","tagline":"Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-4b:free","name":"Qwen: Qwen3 4B (free)","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-4B is a 4 billion parameter dense language model from the Qwen3 series, designed to support both general-purpose and reasoning-intensive tasks.","status":"deprecated","deprecation_date":"2026-03-29","context_window":40960,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"model_type":"chat","release_date":"2025-04-30","tagline":"Qwen3-4B is a 4 billion parameter dense language model from the Qwen3 series, designed to support both general-purpose and reasoning-intensive tasks.","tools":["function_calling"]},{"id":"qwen/qwen3-8b","name":"Qwen: Qwen3 8B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.05,"output":0.4,"cached_input":0.05},"model_type":"chat","parameters":8,"release_date":"2025-04-28","tagline":"Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder-30b-a3b-instruct","name":"Qwen: Qwen3 Coder 30B A3B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.","context_window":160000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.07,"output":0.27},"model_type":"code","parameters":30,"active_parameters":3,"release_date":"2025-07-31","tagline":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, an","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder-flash","name":"Qwen: Qwen3 Coder Flash","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.195,"output":0.975,"cached_input":0.039},"model_type":"code","release_date":"2025-09-17","tagline":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder:free","name":"Qwen: Qwen3 Coder 480B A35B (free)","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","context_window":1048576,"max_output_tokens":262000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"model_type":"code","parameters":480,"active_parameters":35,"release_date":"2025-07-23","tagline":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","tools":["function_calling"]},{"id":"qwen/qwen3-coder-next","name":"Qwen: Qwen3 Coder Next","created_by":"qwen","source":"official","last_updated":"2026-05-07","family":"qwen3","description":"Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.11,"output":0.8,"cached_input":0.07},"model_type":"code","release_date":"2026-02-04","tagline":"Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder-plus","name":"Qwen: Qwen3 Coder Plus","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.65,"output":3.25,"cached_input":0.13},"model_type":"code","release_date":"2025-09-23","tagline":"Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder","name":"Qwen: Qwen3 Coder 480B A35B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","context_window":1048576,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.22,"output":1.8,"cached_input":0.022},"model_type":"code","parameters":480,"active_parameters":35,"release_date":"2025-07-23","tagline":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","tools":["function_calling"],"max_output_tokens":65536},{"id":"qwen/qwen3-max-thinking","name":"Qwen: Qwen3 Max Thinking","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.78,"output":3.9},"model_type":"chat","release_date":"2026-02-09","tagline":"Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning.","tools":["function_calling"]},{"id":"qwen/qwen3-max","name":"Qwen: Qwen3 Max","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.78,"output":3.9,"cached_input":0.156},"model_type":"chat","release_date":"2025-09-23","tagline":"Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the Janua","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-instruct:free","name":"Qwen: Qwen3 Next 80B A3B Instruct (free)","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"model_type":"chat","parameters":80,"active_parameters":3,"release_date":"2025-09-11","tagline":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","tools":["function_calling"]},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"Qwen: Qwen3 Next 80B A3B Instruct","created_by":"qwen","source":"official","last_updated":"2026-04-25","family":"qwen3","description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.09,"output":1.1},"model_type":"chat","parameters":80,"active_parameters":3,"release_date":"2025-09-11","tagline":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","tools":["function_calling"],"max_output_tokens":16384,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"Qwen: Qwen3 Next 80B A3B Thinking","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.098,"output":0.78},"model_type":"chat","parameters":80,"active_parameters":3,"release_date":"2025-09-11","tagline":"Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-235b-a22b-instruct","name":"Qwen: Qwen3 VL 235B A22B Instruct","created_by":"qwen","source":"official","last_updated":"2026-04-25","family":"qwen3","description":"Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.88,"cached_input":0.11},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-09-23","tagline":"Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video.","tools":["function_calling"],"max_output_tokens":16384,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-235b-a22b-thinking","name":"Qwen: Qwen3 VL 235B A22B Thinking","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.26,"output":2.6},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-09-23","tagline":"Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-30b-a3b-instruct","name":"Qwen: Qwen3 VL 30B A3B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.13,"output":0.52},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-10-06","tagline":"Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-30b-a3b-thinking","name":"Qwen: Qwen3 VL 30B A3B Thinking","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.13,"output":1.56},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-10-06","tagline":"Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-32b-instruct","name":"Qwen: Qwen3 VL 32B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.104,"output":0.416},"model_type":"chat","parameters":32,"release_date":"2025-10-23","tagline":"Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-8b-instruct","name":"Qwen: Qwen3 VL 8B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video.","context_window":256000,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.08,"output":0.5},"model_type":"chat","parameters":8,"release_date":"2025-10-14","tagline":"Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-8b-thinking","name":"Qwen: Qwen3 VL 8B Thinking","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences.","context_window":256000,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.117,"output":1.365},"model_type":"chat","parameters":8,"release_date":"2025-10-14","tagline":"Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-122b-a10b","name":"Qwen: Qwen3.5-122B-A10B","created_by":"qwen","source":"official","last_updated":"2026-05-23","family":"qwen3.5","description":"The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.26,"output":2.08},"model_type":"chat","parameters":122,"active_parameters":10,"release_date":"2026-02-25","tagline":"The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference eff","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-27b","name":"Qwen: Qwen3.5-27B","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","description":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","context_window":262144,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.195,"output":1.56},"model_type":"chat","parameters":27,"release_date":"2026-02-25","tagline":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-35b-a3b","name":"Qwen: Qwen3.5-35B-A3B","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3.5","description":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.14,"output":1,"cached_input":0.05},"model_type":"chat","parameters":35,"active_parameters":3,"release_date":"2026-02-25","tagline":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inf","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen: Qwen3.5 397B A17B","created_by":"qwen","source":"official","last_updated":"2026-06-18","family":"qwen3.5","description":"The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":256000,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.385,"output":2.45,"cached_input":0.195},"model_type":"chat","parameters":397,"active_parameters":17,"release_date":"2026-02-16","tagline":"The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher infere","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-9b","name":"Qwen: Qwen3.5-9B","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3.5","description":"Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture.","context_window":262144,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.1,"output":0.15},"model_type":"chat","parameters":9,"release_date":"2026-03-10","tagline":"Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture.","tools":["function_calling"],"max_output_tokens":262144},{"id":"qwen/qwen3.5-flash-02-23","name":"Qwen: Qwen3.5-Flash","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","description":"The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.065,"output":0.26},"model_type":"chat","release_date":"2026-02-25","tagline":"The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference effic","tools":["function_calling"]},{"id":"qwen/qwen3.5-plus-02-15","name":"Qwen: Qwen3.5 Plus 2026-02-15","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","description":"The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference efficiency.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.26,"output":1.56},"model_type":"chat","release_date":"2026-02-16","tagline":"The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference e","tools":["function_calling"]},{"id":"qwen/qwen3.5-plus-20260420","name":"Qwen: Qwen3.5 Plus 2026-04-20","created_by":"qwen","source":"official","family":"qwen3.5","description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","alias":"qwen/qwen3.5-plus","tagline":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.3,"output":1.8},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"qwen/qwen3.5-plus","name":"Qwen: Qwen3.5 Plus 2026-04-20","created_by":"qwen","source":"official","family":"qwen3.5","description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.3,"output":1.8},"tools":["function_calling"],"snapshots":["qwen/qwen3.5-plus-20260420"],"last_updated":"2026-05-17","license":"proprietary","open_weight":false},{"id":"qwen/qwen3.6-27b","name":"Qwen: Qwen3.6 27B","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":262140,"parameters":27,"model_type":"chat","tagline":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.289,"output":3.17},"tools":["function_calling"],"last_updated":"2026-06-18","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen: Qwen3.6 35B A3B","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":262144,"parameters":35,"active_parameters":3,"model_type":"chat","tagline":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"tool_call":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.14,"output":1,"cached_input":0.05},"last_updated":"2026-06-20","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-flash","name":"Qwen: Qwen3.6 Flash","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.188,"output":1.125},"tools":["function_calling"],"last_updated":"2026-05-17","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-max-preview","name":"Qwen: Qwen3.6 Max Preview","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.04,"output":6.24},"tools":["function_calling"],"last_updated":"2026-05-01","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-plus:free","name":"Qwen: Qwen3.6 Plus (free)","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","release_date":"2026-04-02","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"qwen/qwen3.6-plus-preview:free","name":"Qwen: Qwen3.6 Plus Preview (free)","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen 3.6 Plus Preview is the next-generation evolution of the Qwen Plus series, featuring an advanced hybrid architecture that improves efficiency and scalability.","release_date":"2026-03-30","context_window":1000000,"max_output_tokens":32000,"model_type":"chat","tagline":"Qwen 3.6 Plus Preview is the next-generation evolution of the Qwen Plus series, featuring an advanced hybrid architecture that improves efficiency and scalability.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"qwen/qwen3.6-plus","name":"Qwen: Qwen3.6 Plus","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","release_date":"2026-04-02","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.325,"output":1.95},"tools":["function_calling"],"last_updated":"2026-04-08","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.7-max","name":"Qwen: Qwen3.7 Max","created_by":"qwen","source":"official","family":"qwen3.7","description":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series.","release_date":"2026-05-21","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.25,"output":3.75,"cached_input":0.25},"tools":["function_calling"],"last_updated":"2026-05-29","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.7-plus","name":"Qwen: Qwen3.7 Plus","created_by":"qwen","source":"official","family":"qwen3.7","description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series.","release_date":"2026-06-03","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.32,"output":1.28,"cached_input":0.064},"tools":["function_calling"],"last_updated":"2026-06-18","license":"apache-2.0","open_weight":true},{"id":"qwen/qwq-32b","name":"Qwen: QwQ 32B","created_by":"qwen","source":"official","last_updated":"2026-04-22","family":"qwq","description":"QwQ is the reasoning model of the Qwen series.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.58},"model_type":"chat","parameters":32,"release_date":"2025-03-05","tagline":"QwQ is the reasoning model of the Qwen series.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-29","license":"apache-2.0","open_weight":true},{"id":"reka/reka-edge","name":"Reka Edge","created_by":"reka","source":"official","description":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","release_date":"2026-03-20","context_window":16384,"max_output_tokens":16384,"tagline":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"vision":true,"streaming":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"tools":["function_calling"],"last_updated":"2026-03-27","license":"proprietary","open_weight":false},{"id":"rekaai/reka-edge","name":"Reka Edge","created_by":"rekaai","source":"official","description":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","release_date":"2026-03-20","context_window":16384,"max_output_tokens":16384,"tagline":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"rekaai/reka-flash-3","name":"Reka Flash 3","created_by":"rekaai","source":"official","description":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka.","release_date":"2025-03-12","context_window":65536,"max_output_tokens":65536,"tagline":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka.","capabilities":{"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.2},"last_updated":"2026-06-20"},{"id":"relace/relace-apply-3","name":"Relace: Relace Apply 3","created_by":"relace","source":"official","last_updated":"2026-03-23","description":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files.","context_window":256000,"max_output_tokens":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.85,"output":1.25},"release_date":"2025-09-26","tagline":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files."},{"id":"relace/relace-search","name":"Relace: Relace Search","created_by":"relace","source":"official","last_updated":"2026-06-20","description":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request.","context_window":256000,"max_output_tokens":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1,"output":3},"model_type":"chat","release_date":"2025-12-08","tagline":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request.","tools":["function_calling"]},{"id":"sao10k/l3-euryale-70b","name":"Sao10k: Llama 3 Euryale 70B v2.1","created_by":"sao10k","source":"official","last_updated":"2026-05-29","description":"Euryale 70B v2.1 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","context_window":8192,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":1.48,"output":1.48},"parameters":70,"release_date":"2024-06-18","tagline":"Euryale 70B v2.1 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-06-05"},{"id":"sao10k/l3-lunaris-8b","name":"Sao10K: Llama 3 8B Lunaris","created_by":"sao10k","source":"official","last_updated":"2026-04-25","description":"Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3.","context_window":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.04,"output":0.05},"parameters":8,"release_date":"2024-08-13","tagline":"Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3.","max_output_tokens":16384},{"id":"sao10k/l3.1-70b-hanami-x1","name":"Sao10K: Llama 3.1 70B Hanami x1","created_by":"sao10k","source":"official","last_updated":"2026-06-20","description":"This is [Sao10K](/sao10k)'s experiment over [Euryale v2.2](/sao10k/l3.1-euryale-70b).","context_window":16000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":3,"output":3},"parameters":70,"release_date":"2025-01-08","tagline":"This is [Sao10K](/sao10k)'s experiment over [Euryale v2.2](/sao10k/l3.1-euryale-70b).","capabilities":{"structured_output":true,"json_mode":true}},{"id":"sao10k/l3.1-euryale-70b","name":"Sao10K: Llama 3.1 Euryale 70B v2.2","created_by":"sao10k","source":"official","last_updated":"2026-03-24","description":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.85,"output":0.85},"parameters":70,"release_date":"2024-08-28","tagline":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","tools":["function_calling"]},{"id":"sao10k/l3.3-euryale-70b","name":"Sao10K: Llama 3.3 Euryale 70B","created_by":"sao10k","source":"official","last_updated":"2026-03-24","description":"Euryale L3.3 70B is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.65,"output":0.75},"parameters":70,"release_date":"2024-12-18","tagline":"Euryale L3.3 70B is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k)."},{"id":"stepfun/step-3.5-flash:free","name":"StepFun: Step 3.5 Flash (free)","created_by":"stepfun","source":"official","last_updated":"2026-03-23","description":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","context_window":256000,"max_output_tokens":256000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"release_date":"2026-01-29","tagline":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","tools":["function_calling"]},{"id":"stepfun/step-3.5-flash","name":"StepFun: Step 3.5 Flash","created_by":"stepfun","source":"official","last_updated":"2026-05-23","description":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","context_window":262144,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.09,"output":0.3,"cached_input":0.02},"release_date":"2026-01-29","tagline":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"stepfun/step-3.7-flash","name":"StepFun: Step 3.7 Flash","created_by":"stepfun","source":"official","description":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model.","release_date":"2026-05-28","context_window":256000,"max_output_tokens":256000,"tagline":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.2,"output":1.15,"cached_input":0.04},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"switchpoint/router","name":"Switchpoint Router","created_by":"switchpoint","source":"official","last_updated":"2026-06-20","description":"Switchpoint AI's router instantly analyzes your request and directs it to the optimal AI from an ever-evolving library.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.85,"output":3.4},"release_date":"2025-07-11","tagline":"Switchpoint AI's router instantly analyzes your request and directs it to the optimal AI from an ever-evolving library."},{"id":"tencent/hunyuan-a13b-instruct","name":"Tencent: Hunyuan A13B Instruct","created_by":"tencent","source":"official","last_updated":"2026-03-24","description":"Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.14,"output":0.57},"release_date":"2025-07-08","tagline":"Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought."},{"id":"tencent/hy3-preview:free","name":"Tencent: Hy3 preview (free)","created_by":"tencent","source":"official","description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","release_date":"2026-04-22","context_window":262144,"max_output_tokens":262144,"tagline":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","status":"deprecated","deprecation_date":"2026-05-08"},{"id":"tencent/hy3-preview","name":"Tencent: Hy3 preview","created_by":"tencent","source":"official","description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","release_date":"2026-04-22","context_window":262144,"max_output_tokens":262144,"tagline":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.063,"output":0.21,"cached_input":0.021},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"thedrummer/cydonia-24b-v4.1","name":"TheDrummer: Cydonia 24B V4.1","created_by":"thedrummer","source":"official","last_updated":"2026-06-18","description":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.5,"cached_input":0.15},"parameters":24,"release_date":"2025-09-27","tagline":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"thedrummer/rocinante-12b","name":"TheDrummer: Rocinante 12B","created_by":"thedrummer","source":"official","last_updated":"2026-03-24","description":"Rocinante 12B is designed for engaging storytelling and rich prose.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.17,"output":0.43},"parameters":12,"release_date":"2024-09-30","tagline":"Rocinante 12B is designed for engaging storytelling and rich prose.","tools":["function_calling"]},{"id":"thedrummer/skyfall-36b-v2","name":"TheDrummer: Skyfall 36B V2","created_by":"thedrummer","source":"official","last_updated":"2026-06-20","description":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":0.8,"cached_input":0.25},"parameters":36,"release_date":"2025-03-10","tagline":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"thedrummer/unslopnemo-12b","name":"TheDrummer: UnslopNemo 12B","created_by":"thedrummer","source":"official","last_updated":"2026-03-24","description":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":0.4},"parameters":12,"release_date":"2024-11-08","tagline":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","tools":["function_calling"]},{"id":"tngtech/deepseek-r1t2-chimera","name":"TNG: DeepSeek R1T2 Chimera","created_by":"tngtech","source":"official","last_updated":"2026-03-24","family":"deepseek-r1","description":"DeepSeek-TNG-R1T2-Chimera is the second-generation Chimera model from TNG Tech.","context_window":163840,"max_output_tokens":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.3,"output":1.1,"cached_input":0.15},"model_type":"reasoning","release_date":"2025-07-08","tagline":"DeepSeek-TNG-R1T2-Chimera is the second-generation Chimera model from TNG Tech.","tools":["function_calling"]},{"id":"undi95/remm-slerp-l2-13b","name":"ReMM SLERP 13B","created_by":"undi95","source":"official","last_updated":"2026-03-24","description":"A recreation trial of the original MythoMax-L2-B13 but with updated models.","context_window":6144,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.45,"output":0.65},"parameters":13,"release_date":"2023-07-22","tagline":"A recreation trial of the original MythoMax-L2-B13 but with updated models."},{"id":"upstage/solar-pro-3","name":"Upstage: Solar Pro 3","created_by":"upstage","source":"official","last_updated":"2026-03-24","description":"Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.015},"release_date":"2026-01-27","tagline":"Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model.","tools":["function_calling"]},{"id":"writer/palmyra-x5","name":"Writer: Palmyra X5","created_by":"writer","source":"official","last_updated":"2026-03-23","description":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise.","context_window":1040000,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":6},"model_type":"chat","release_date":"2026-01-21","tagline":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise.","license":"proprietary","open_weight":false,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true}},{"id":"x-ai/grok-3-beta","name":"xAI: Grok 3 Beta","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"Grok 3 is the latest model from xAI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.75},"model_type":"chat","release_date":"2025-04-09","tagline":"Grok 3 is the latest model from xAI.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-3-mini-beta","name":"xAI: Grok 3 Mini Beta","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"Grok 3 Mini is a lightweight, smaller thinking model.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.3,"output":0.5,"cached_input":0.075},"model_type":"chat","release_date":"2025-04-09","tagline":"Grok 3 Mini is a lightweight, smaller thinking model.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-3-mini","name":"xAI: Grok 3 Mini","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"A lightweight model that thinks before responding.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true},"pricing":{"input":0.3,"output":0.5,"cached_input":0.075},"model_type":"chat","release_date":"2025-06-10","tagline":"A lightweight model that thinks before responding.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"x-ai/grok-3","name":"xAI: Grok 3","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"Grok 3 is the latest model from xAI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.75},"model_type":"chat","release_date":"2025-06-10","tagline":"Grok 3 is the latest model from xAI.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"x-ai/grok-4-fast","name":"xAI: Grok 4 Fast","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-4","description":"Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window.","context_window":2000000,"max_output_tokens":30000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5,"cached_input":0.05},"model_type":"chat","release_date":"2025-09-19","tagline":"Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-4.1-fast","name":"xAI: Grok 4.1 Fast","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-4.1","description":"Grok 4.1 Fast is xAI's best agentic tool calling model that shines in real-world use cases like customer support and deep research.","context_window":2000000,"max_output_tokens":30000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5,"cached_input":0.05},"model_type":"chat","release_date":"2025-11-19","tagline":"Grok 4.1 Fast is xAI's best agentic tool calling model that shines in real-world use cases like customer support and deep research.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-4.20-beta","name":"xAI: Grok 4.20 Beta","created_by":"xai","source":"official","last_updated":"2026-03-24","family":"grok-4.20","description":"Grok 4.20 Beta is xAI's newest flagship model with industry-leading speed and agentic tool calling capabilities.","context_window":2000000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2026-03-12","tagline":"Grok 4.20 Beta is xAI's newest flagship model with industry-leading speed and agentic tool calling capabilities.","tools":["function_calling"]},{"id":"x-ai/grok-4.20-multi-agent-beta","name":"xAI: Grok 4.20 Multi-Agent Beta","created_by":"xai","source":"official","last_updated":"2026-03-24","family":"grok-4.20","description":"Grok 4.20 Multi-Agent Beta is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows.","context_window":2000000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2026-03-12","tagline":"Grok 4.20 Multi-Agent Beta is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows."},{"id":"x-ai/grok-4.20-multi-agent","name":"xAI: Grok 4.20 Multi-Agent","created_by":"xai","source":"official","family":"grok-4.20","description":"Grok 4.20 Multi-Agent is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows.","release_date":"2026-03-31","context_window":2000000,"model_type":"chat","tagline":"Grok 4.20 Multi-Agent is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"last_updated":"2026-06-18"},{"id":"x-ai/grok-4.20","name":"xAI: Grok 4.20","created_by":"xai","source":"official","family":"grok-4.20","description":"Grok 4.20 is a reasoning model from xAI with industry-leading speed and agentic tool calling capabilities.","release_date":"2026-03-31","context_window":2000000,"model_type":"chat","tagline":"Grok 4.20 is xAI's newest flagship model with industry-leading speed and agentic tool calling capabilities.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-10"},{"id":"x-ai/grok-4.3","name":"xAI: Grok 4.3","created_by":"xai","source":"official","family":"grok-4.3","description":"Grok 4.3 is a reasoning model from xAI.","release_date":"2026-04-30","context_window":1000000,"model_type":"chat","tagline":"Grok 4.3 is a reasoning model from xAI.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-01","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"x-ai/grok-4","name":"xAI: Grok 4","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-4","description":"Grok 4 is xAI's latest reasoning model with a 256k context window.","context_window":256000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.75},"model_type":"chat","release_date":"2025-07-09","tagline":"Grok 4 is xAI's latest reasoning model with a 256k context window.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-build-0.1","name":"xAI: Grok Build 0.1","created_by":"xai","source":"official","description":"Grok Build 0.1 is xAI’s fast coding model trained specifically for agentic software engineering workflows.","release_date":"2026-05-20","context_window":256000,"model_type":"chat","tagline":"Grok Build 0.1 is xAI’s fast coding model trained specifically for agentic software engineering workflows.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1,"output":2,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-23","license":"proprietary","open_weight":false},{"id":"x-ai/grok-code-fast-1","name":"xAI: Grok Code Fast 1","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-code","description":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","context_window":256000,"max_output_tokens":10000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true},"pricing":{"input":0.2,"output":1.5,"cached_input":0.02},"model_type":"code","release_date":"2025-08-26","tagline":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15","license":"proprietary","open_weight":false},{"id":"xiaomi/mimo-v2-flash","name":"Xiaomi: MiMo-V2-Flash","created_by":"xiaomi","source":"official","last_updated":"2026-05-10","description":"MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi.","context_window":262144,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"release_date":"2025-12-14","tagline":"MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi.","tools":["function_calling"],"family":"mimo","model_type":"chat","knowledge_cutoff":"2024-12","license":"mit","parameters":309,"active_parameters":15,"architecture":"moe","open_weight":true},{"id":"xiaomi/mimo-v2-omni","name":"Xiaomi: MiMo-V2-Omni","created_by":"xiaomi","source":"official","last_updated":"2026-05-29","description":"MiMo-V2-Omni is a frontier omni-modal model that natively processes image, video, and audio inputs within a unified architecture.","context_window":262144,"max_output_tokens":65536,"modalities":{"input":["text","audio","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.4,"output":2,"cached_input":0.08},"release_date":"2026-03-18","tagline":"MiMo-V2-Omni is a frontier omni-modal model that natively processes image, video, and audio inputs within a unified architecture.","tools":["function_calling"],"family":"mimo","model_type":"chat","status":"deprecated","deprecation_date":"2026-05-31","license":"proprietary","open_weight":false},{"id":"xiaomi/mimo-v2-pro","name":"Xiaomi: MiMo-V2-Pro","created_by":"xiaomi","source":"official","last_updated":"2026-05-29","description":"MiMo-V2-Pro is Xiaomi's flagship foundation model, featuring over 1T total parameters and a 1M context length, deeply optimized for agentic scenarios.","context_window":1048576,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":1,"output":3,"cached_input":0.2},"release_date":"2026-03-18","tagline":"MiMo-V2-Pro is Xiaomi's flagship foundation model, featuring over 1T total parameters and a 1M context length, deeply optimized for agentic scenarios.","tools":["function_calling"],"family":"mimo","model_type":"chat","status":"deprecated","deprecation_date":"2026-05-31","license":"proprietary","parameters":1000,"active_parameters":42,"architecture":"moe","open_weight":false},{"id":"xiaomi/mimo-v2.5-pro","name":"Xiaomi: MiMo-V2.5-Pro","created_by":"xiaomi","source":"official","family":"mimo","description":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as C","release_date":"2026-04-22","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as C","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.435,"output":0.87,"cached_input":0.004},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"xiaomi/mimo-v2.5","name":"Xiaomi: MiMo-V2.5","created_by":"xiaomi","source":"official","family":"mimo","description":"MiMo-V2.5 is a native omnimodal model by Xiaomi.","release_date":"2026-04-22","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"MiMo-V2.5 is a native omnimodal model by Xiaomi.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"modalities":{"input":["text","audio","image","video"],"output":["text"]},"pricing":{"input":0.14,"output":0.28,"cached_input":0.003},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"z-ai/glm-4-32b","name":"Z.ai: GLM 4 32B ","created_by":"z-ai","source":"official","last_updated":"2026-03-23","description":"GLM 4 32B is a cost-effective foundation language model.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.1,"output":0.1},"family":"glm-4","model_type":"chat","parameters":32,"release_date":"2025-07-24","tagline":"GLM 4 32B is a cost-effective foundation language model.","tools":["function_calling"]},{"id":"z-ai/glm-4.5-air:free","name":"Z.ai: GLM 4.5 Air (free)","created_by":"z-ai","source":"official","last_updated":"2026-03-23","description":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","context_window":131072,"max_output_tokens":96000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"family":"glm-4.5","model_type":"chat","release_date":"2025-07-25","tagline":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","tools":["function_calling"]},{"id":"z-ai/glm-4.5-air","name":"Z.ai: GLM 4.5 Air","created_by":"z-ai","source":"official","last_updated":"2026-06-18","description":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","context_window":131072,"max_output_tokens":98304,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.13,"output":0.85,"cached_input":0.025},"family":"glm-4.5","model_type":"chat","release_date":"2025-07-25","tagline":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","tools":["function_calling"]},{"id":"z-ai/glm-4.5","name":"Z.ai: GLM 4.5","created_by":"z-ai","source":"official","last_updated":"2026-06-20","description":"GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications.","context_window":131072,"max_output_tokens":98304,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":2.2,"cached_input":0.11},"family":"glm-4.5","model_type":"chat","release_date":"2025-07-25","tagline":"GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-12-31"},{"id":"z-ai/glm-4.5v","name":"Z.ai: GLM 4.5V","created_by":"z-ai","source":"official","last_updated":"2026-03-24","description":"GLM-4.5V is a vision-language foundation model for multimodal agent applications.","context_window":65536,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":1.8,"cached_input":0.11},"family":"glm-4.5","model_type":"chat","release_date":"2025-08-11","tagline":"GLM-4.5V is a vision-language foundation model for multimodal agent applications.","tools":["function_calling"]},{"id":"z-ai/glm-4.6","name":"Z.ai: GLM 4.6","created_by":"z-ai","source":"official","last_updated":"2026-05-14","description":"Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex.","context_window":202752,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.43,"output":1.74,"cached_input":0.08},"family":"glm-4.6","model_type":"chat","release_date":"2025-09-30","tagline":"Compared with GLM-4.5, this generation brings several key improvements:\n\nLonger context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-14"},{"id":"z-ai/glm-4.6v","name":"Z.ai: GLM 4.6V","created_by":"z-ai","source":"official","last_updated":"2026-06-10","description":"GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.3,"output":0.9,"cached_input":0.055},"family":"glm-4.6","model_type":"chat","release_date":"2025-12-08","tagline":"GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media.","tools":["function_calling"]},{"id":"z-ai/glm-4.7-flash","name":"Z.ai: GLM 4.7 Flash","created_by":"z-ai","source":"official","last_updated":"2026-04-25","description":"As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency.","context_window":202752,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.06,"output":0.4,"cached_input":0.01},"family":"glm-4.7","model_type":"chat","release_date":"2026-01-19","tagline":"As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency.","tools":["function_calling"],"max_output_tokens":16384},{"id":"z-ai/glm-4.7","name":"Z.ai: GLM 4.7","created_by":"z-ai","source":"official","last_updated":"2026-05-10","description":"GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution.","context_window":202752,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.4,"output":1.75,"cached_input":0.08},"family":"glm-4.7","model_type":"chat","release_date":"2025-12-22","tagline":"GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution.","tools":["function_calling"]},{"id":"z-ai/glm-5-turbo","name":"Z.ai: GLM 5 Turbo","created_by":"z-ai","source":"official","last_updated":"2026-06-10","description":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios.","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.2,"output":4,"cached_input":0.24},"family":"glm-5","model_type":"chat","release_date":"2026-03-15","tagline":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios.","tools":["function_calling"]},{"id":"z-ai/glm-5.1","name":"Z.ai: GLM 5.1","created_by":"z-ai","source":"official","family":"glm-5.1","description":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.","release_date":"2026-04-07","context_window":202752,"model_type":"chat","tagline":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.98,"output":3.08,"cached_input":0.182},"tools":["function_calling"],"last_updated":"2026-06-02","max_output_tokens":131072},{"id":"z-ai/glm-5.2","name":"Z.ai: GLM 5.2","created_by":"z-ai","source":"official","family":"glm-5.2","description":"GLM 5.2 is a large-scale reasoning model from Z.ai.","release_date":"2026-06-16","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM 5.2 is a large-scale reasoning model from Z.ai.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.2,"output":4.1,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"z-ai/glm-5","name":"Z.ai: GLM 5","created_by":"z-ai","source":"official","last_updated":"2026-06-03","description":"GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows.","context_window":202752,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":1.92,"cached_input":0.12},"family":"glm-5","model_type":"chat","release_date":"2026-02-11","tagline":"GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows.","tools":["function_calling"]},{"id":"z-ai/glm-5v-turbo","name":"Z.ai: GLM 5V Turbo","created_by":"z-ai","source":"official","family":"glm-5","description":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks.","release_date":"2026-04-01","context_window":202752,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":1.2,"output":4,"cached_input":0.24},"tools":["function_calling"],"last_updated":"2026-04-03"}]};
+export const provider: ProviderWithModels = {"id":"openrouter","name":"OpenRouter","region":"US","headquarters":"New York, NY","founded":2023,"url":"https://openrouter.ai","api_url":"https://openrouter.ai/api/v1","docs_url":"https://openrouter.ai/docs","pricing_url":"https://openrouter.ai/models","description":"Unified API gateway providing access to 200+ models from multiple providers with standardized pricing.","type":"aggregator","playground_url":"https://openrouter.ai/playground","status_url":"https://status.openrouter.ai","openai_compatible":true,"free_tier":true,"github_url":"https://github.com/OpenRouterTeam","models_url":"https://openrouter.ai/models","twitter_url":"https://x.com/OpenRouterAI","discord_url":"https://discord.com/invite/openrouter","terms_url":"https://openrouter.ai/terms","support_url":"https://openrouter.ai/support","icon":"","models":[{"id":"~anthropic/claude-fable-latest","name":"Anthropic: Claude Fable Latest","created_by":"~anthropic","source":"official","family":"claude-fable","description":"This model always redirects to the latest model in the Claude Fable family.","release_date":"2026-06-09","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Claude Fable family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":10,"output":50,"cached_input":1},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"~anthropic/claude-haiku-latest","name":"Anthropic Claude Haiku Latest","created_by":"~anthropic","source":"official","family":"claude-haiku","description":"This model always redirects to the latest model in the Anthropic Claude Haiku family.","release_date":"2026-04-27","context_window":200000,"max_output_tokens":64000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Anthropic Claude Haiku family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1,"output":5,"cached_input":0.1},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"~anthropic/claude-opus-latest","name":"Anthropic: Claude Opus Latest","created_by":"~anthropic","source":"official","family":"claude-opus","description":"This model always redirects to the latest model in the Claude Opus family.","release_date":"2026-04-21","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Claude Opus family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":25,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-04-22"},{"id":"~anthropic/claude-sonnet-latest","name":"Anthropic Claude Sonnet Latest","created_by":"~anthropic","source":"official","family":"claude-sonnet","description":"This model always redirects to the latest model in the Anthropic Claude Sonnet family.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"This model always redirects to the latest model in the Anthropic Claude Sonnet family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":3,"output":15,"cached_input":0.3},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"~google/gemini-flash-latest","name":"Google Gemini Flash Latest","created_by":"~google","source":"official","description":"This model always redirects to the latest model in the Google Gemini Flash family.","release_date":"2026-04-27","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"This model always redirects to the latest model in the Google Gemini Flash family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"pricing":{"input":1.5,"output":9,"cached_input":0.15},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"~google/gemini-pro-latest","name":"Google Gemini Pro Latest","created_by":"~google","source":"official","description":"This model always redirects to the latest model in the Google Gemini Pro family.","release_date":"2026-04-27","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"This model always redirects to the latest model in the Google Gemini Pro family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["audio","image","text","video"],"output":["text"]},"pricing":{"input":2,"output":12,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"~moonshotai/kimi-latest","name":"MoonshotAI Kimi Latest","created_by":"~moonshotai","source":"official","family":"kimi","description":"This model always redirects to the latest model in the MoonshotAI Kimi family.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"This model always redirects to the latest model in the MoonshotAI Kimi family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.55,"output":3.2,"cached_input":0.11},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"~openai/gpt-latest","name":"OpenAI GPT Latest","created_by":"~openai","source":"official","description":"This model always redirects to the latest model in the OpenAI GPT family.","release_date":"2026-04-27","context_window":1050000,"max_output_tokens":128000,"tagline":"This model always redirects to the latest model in the OpenAI GPT family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"~openai/gpt-mini-latest","name":"OpenAI GPT Mini Latest","created_by":"~openai","source":"official","description":"This model always redirects to the latest model in the OpenAI GPT Mini family.","release_date":"2026-04-27","context_window":400000,"max_output_tokens":128000,"tagline":"This model always redirects to the latest model in the OpenAI GPT Mini family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.75,"output":4.5,"cached_input":0.075},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"ai21/jamba-large-1.7","name":"AI21: Jamba Large 1.7","created_by":"ai21","source":"official","last_updated":"2026-03-24","description":"Jamba Large 1.7 is the latest model in the Jamba open family, offering improvements in grounding, instruction-following, and overall efficiency.","context_window":256000,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":2,"output":8},"model_type":"chat","release_date":"2025-08-08","tagline":"Jamba Large 1.7 is the latest model in the Jamba open family, offering improvements in grounding, instruction-following, and overall efficiency.","tools":["function_calling"]},{"id":"aion-labs/aion-1.0-mini","name":"AionLabs: Aion-1.0-Mini","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-1.0-Mini 32B parameter model is a distilled version of the DeepSeek-R1 model, designed for strong performance in reasoning domains such as mathematics, coding, and logic.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.7,"output":1.4},"release_date":"2025-02-04","tagline":"Aion-1.0-Mini 32B parameter model is a distilled version of the DeepSeek-R1 model, designed for strong performance in reasoning domains such as mathematics, coding, and logic."},{"id":"aion-labs/aion-1.0","name":"AionLabs: Aion-1.0","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-1.0 is a multi-model system designed for high performance across various tasks, including reasoning and coding.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":4,"output":8},"release_date":"2025-02-04","tagline":"Aion-1.0 is a multi-model system designed for high performance across various tasks, including reasoning and coding."},{"id":"aion-labs/aion-2.0","name":"AionLabs: Aion-2.0","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.8,"output":1.6,"cached_input":0.2},"release_date":"2026-02-23","tagline":"Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling."},{"id":"aion-labs/aion-rp-llama-3.1-8b","name":"AionLabs: Aion-RP 1.0 (8B)","created_by":"aion-labs","source":"official","last_updated":"2026-03-23","description":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.6},"parameters":8,"release_date":"2025-02-04","tagline":"Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses."},{"id":"alfredpros/codellama-7b-instruct-solidity","name":"AlfredPros: CodeLLaMa 7B Instruct Solidity","created_by":"alfredpros","source":"official","last_updated":"2026-03-23","description":"A finetuned 7 billion parameters Code LLaMA - Instruct model to generate Solidity smart contract using 4-bit QLoRA finetuning provided by PEFT library.","context_window":4096,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.2},"family":"codellama","model_type":"code","parameters":7,"release_date":"2025-04-14","tagline":"A finetuned 7 billion parameters Code LLaMA - Instruct model to generate Solidity smart contract using 4-bit QLoRA finetuning provided by PEFT library."},{"id":"alibaba/tongyi-deepresearch-30b-a3b","name":"Tongyi DeepResearch 30B A3B","created_by":"alibaba","source":"official","last_updated":"2026-03-24","description":"Tongyi DeepResearch is an agentic large language model developed by Tongyi Lab, with 30 billion total parameters activating only 3 billion per token.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.09,"output":0.45,"cached_input":0.09},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-09-18","tagline":"Tongyi DeepResearch is an agentic large language model developed by Tongyi Lab, with 30 billion total parameters activating only 3 billion per token.","tools":["function_calling"]},{"id":"allenai/molmo-2-8b","name":"AllenAI: Molmo2 8B","created_by":"allenai","source":"official","last_updated":"2026-03-21","description":"Molmo2-8B is an open vision-language model developed by the Allen Institute for AI (Ai2) as part of the Molmo2 family, supporting image, video, and multi-image understanding and grounding.","status":"deprecated","deprecation_date":"2026-03-23","context_window":36864,"max_output_tokens":36864,"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"parameters":8},{"id":"allenai/olmo-2-0325-32b-instruct","name":"AllenAI: Olmo 2 32B Instruct","created_by":"allenai","source":"official","last_updated":"2026-03-23","description":"OLMo-2 32B Instruct is a supervised instruction-finetuned variant of the OLMo-2 32B March 2025 base model.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.2},"parameters":32,"release_date":"2025-03-14","tagline":"OLMo-2 32B Instruct is a supervised instruction-finetuned variant of the OLMo-2 32B March 2025 base model."},{"id":"allenai/olmo-3-32b-think","name":"AllenAI: Olmo 3 32B Think","created_by":"allenai","source":"official","last_updated":"2026-03-24","description":"Olmo 3 32B Think is a large-scale, 32-billion-parameter model purpose-built for deep reasoning, complex logic chains and advanced instruction-following scenarios.","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.15,"output":0.5},"parameters":32,"release_date":"2025-11-21","tagline":"Olmo 3 32B Think is a large-scale, 32-billion-parameter model purpose-built for deep reasoning, complex logic chains and advanced instruction-following scenarios."},{"id":"allenai/olmo-3-7b-instruct","name":"AllenAI: Olmo 3 7B Instruct","created_by":"allenai","source":"official","last_updated":"2026-03-21","description":"Olmo 3 7B Instruct is a supervised instruction-fine-tuned variant of the Olmo 3 7B base model, optimized for instruction-following, question-answering, and natural conversational dialogue.","status":"deprecated","deprecation_date":"2026-03-23","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.2},"parameters":7},{"id":"allenai/olmo-3-7b-think","name":"AllenAI: Olmo 3 7B Think","created_by":"allenai","source":"official","last_updated":"2026-03-21","description":"Olmo 3 7B Think is a research-oriented language model in the Olmo family designed for advanced reasoning and instruction-driven tasks.","status":"deprecated","deprecation_date":"2026-03-23","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.12,"output":0.2},"parameters":7},{"id":"allenai/olmo-3.1-32b-instruct","name":"AllenAI: Olmo 3.1 32B Instruct","created_by":"allenai","source":"official","last_updated":"2026-05-01","description":"Olmo 3.1 32B Instruct is a large-scale, 32-billion-parameter instruction-tuned language model engineered for high-performance conversational AI, multi-turn dialogue, and practical instruction following.","context_window":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.6},"parameters":32,"release_date":"2026-01-06","tagline":"Olmo 3.1 32B Instruct is a large-scale, 32-billion-parameter instruction-tuned language model engineered for high-performance conversational AI, multi-turn dialogue, and practical instruction followin","tools":["function_calling"],"max_output_tokens":16384,"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"allenai/olmo-3.1-32b-think","name":"AllenAI: Olmo 3.1 32B Think","created_by":"allenai","source":"official","last_updated":"2026-03-24","description":"Olmo 3.1 32B Think is a large-scale, 32-billion-parameter model designed for deep reasoning, complex multi-step logic, and advanced instruction following.","context_window":65536,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.15,"output":0.5},"parameters":32,"status":"deprecated","release_date":"2025-12-16","deprecation_date":"2026-04-06","tagline":"Olmo 3.1 32B Think is a large-scale, 32-billion-parameter model designed for deep reasoning, complex multi-step logic, and advanced instruction following."},{"id":"alpindale/goliath-120b","name":"Goliath 120B","created_by":"alpindale","source":"official","last_updated":"2026-03-24","description":"A large LLM created by combining two fine-tuned Llama 70B models into one 120B model.","context_window":6144,"max_output_tokens":1024,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":3.75,"output":7.5},"parameters":120,"release_date":"2023-11-10","tagline":"A large LLM created by combining two fine-tuned Llama 70B models into one 120B model."},{"id":"amazon/nova-2-lite-v1","name":"Amazon: Nova 2 Lite","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text.","context_window":1000000,"max_output_tokens":65535,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","release_date":"2025-12-02","tagline":"Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text.","tools":["function_calling"]},{"id":"amazon/nova-lite-v1","name":"Amazon: Nova Lite 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output.","context_window":300000,"max_output_tokens":5120,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.06,"output":0.24},"model_type":"chat","release_date":"2024-12-05","tagline":"Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output.","tools":["function_calling"]},{"id":"amazon/nova-micro-v1","name":"Amazon: Nova Micro 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost.","context_window":128000,"max_output_tokens":5120,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.035,"output":0.14},"model_type":"chat","release_date":"2024-12-05","tagline":"Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost.","tools":["function_calling"]},{"id":"amazon/nova-premier-v1","name":"Amazon: Nova Premier 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","context_window":1000000,"max_output_tokens":32000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":2.5,"output":12.5,"cached_input":0.625},"model_type":"chat","release_date":"2025-10-31","tagline":"Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.","tools":["function_calling"]},{"id":"amazon/nova-pro-v1","name":"Amazon: Nova Pro 1.0","created_by":"amazon","source":"official","last_updated":"2026-03-23","description":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks.","context_window":300000,"max_output_tokens":5120,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.8,"output":3.2},"model_type":"chat","release_date":"2024-12-05","tagline":"Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks.","tools":["function_calling"]},{"id":"anthracite-org/magnum-v4-72b","name":"Magnum v4 72B","created_by":"anthracite-org","source":"official","last_updated":"2026-05-17","description":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus).","context_window":32768,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":3,"output":5},"parameters":72,"release_date":"2024-10-22","tagline":"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthrop"},{"id":"anthropic/claude-3-haiku","name":"Anthropic: Claude 3 Haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","description":"Claude 3 Haiku is Anthropic's fastest and most compact model for\nnear-instant responsiveness. Quick and accurate targeted performance.\n\nSee the launch announcement and benchmark results [here](https:/","context_window":200000,"max_output_tokens":4096,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"streaming":true,"vision":true,"batch":true},"pricing":{"input":0.25,"output":1.25,"cached_input":0.03},"model_type":"chat","release_date":"2024-03-13","tagline":"Claude 3 Haiku is Anthropic's fastest and most compact model for\nnear-instant responsiveness. Quick and accurate targeted performance.\n\nSee the launch announcement and benchmark results [here](https:/","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"anthropic/claude-3.5-haiku","name":"Anthropic: Claude 3.5 Haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","description":"Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use.","context_window":200000,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.8,"output":4,"cached_input":0.08},"model_type":"chat","release_date":"2024-11-04","tagline":"Claude 3.5 Haiku features offers enhanced capabilities in speed, coding accuracy, and tool use.","tools":["function_calling"]},{"id":"anthropic/claude-3.5-sonnet","name":"Anthropic: Claude 3.5 Sonnet","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","description":"New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices.","context_window":200000,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":6,"output":30,"cached_input":0.6},"model_type":"chat","release_date":"2024-10-22","tagline":"New Claude 3.5 Sonnet delivers better-than-Opus capabilities, faster-than-Sonnet speeds, at the same Sonnet prices.","tools":["function_calling"]},{"id":"anthropic/claude-3.7-sonnet:thinking","name":"Anthropic: Claude 3.7 Sonnet (thinking)","created_by":"anthropic","source":"official","last_updated":"2026-05-06","family":"claude-sonnet","description":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","status":"deprecated","deprecation_date":"2026-05-11","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-02-24","tagline":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","tools":["function_calling"]},{"id":"anthropic/claude-3.7-sonnet","name":"Anthropic: Claude 3.7 Sonnet","created_by":"anthropic","source":"official","last_updated":"2026-05-06","family":"claude-sonnet","description":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","status":"deprecated","deprecation_date":"2026-05-11","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-02-24","tagline":"Claude 3.7 Sonnet is an advanced large language model with improved reasoning, coding, and problem-solving capabilities.","tools":["function_calling"]},{"id":"anthropic/claude-fable-5","name":"Anthropic: Claude Fable 5","created_by":"anthropic","source":"official","family":"claude-fable","description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding.","release_date":"2026-06-09","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":10,"output":50,"cached_input":1},"tools":["function_calling"],"last_updated":"2026-06-18","license":"proprietary","open_weight":false},{"id":"anthropic/claude-haiku-4.5","name":"Anthropic: Claude Haiku 4.5","created_by":"anthropic","source":"official","last_updated":"2026-05-14","family":"claude-haiku","description":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models.","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1,"output":5,"cached_input":0.1},"model_type":"chat","release_date":"2025-10-15","tagline":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.1","name":"Anthropic: Claude Opus 4.1","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-opus","description":"Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks.","context_window":200000,"max_output_tokens":32000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":15,"output":75,"cached_input":1.5},"model_type":"chat","release_date":"2025-08-05","tagline":"Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.5","name":"Anthropic: Claude Opus 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-opus","description":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use.","context_window":200000,"max_output_tokens":64000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","release_date":"2025-11-24","tagline":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.6-fast","name":"Anthropic: Claude Opus 4.6 (Fast)","created_by":"anthropic","source":"official","family":"claude-opus","description":"Fast-mode variant of [Opus 4.6](/anthropic/claude-opus-4.6) - identical capabilities with higher output speed at premium 6x pricing.","release_date":"2026-04-07","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Fast-mode variant of [Opus 4.6](/anthropic/claude-opus-4.6) - identical capabilities with higher output speed at premium 6x pricing.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":30,"output":150,"cached_input":3},"tools":["function_calling"],"last_updated":"2026-05-29","status":"deprecated","deprecation_date":"2026-06-29"},{"id":"anthropic/claude-opus-4.6","name":"Anthropic: Claude Opus 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-opus","description":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks.","context_window":1000000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","release_date":"2026-02-04","tagline":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks.","tools":["function_calling"]},{"id":"anthropic/claude-opus-4.7-fast","name":"Anthropic: Claude Opus 4.7 (Fast)","created_by":"anthropic","source":"official","family":"claude-opus","description":"Fast-mode variant of [Opus 4.7](/anthropic/claude-opus-4.7) - identical capabilities with higher output speed at premium 6x pricing.","release_date":"2026-05-12","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Fast-mode variant of [Opus 4.7](/anthropic/claude-opus-4.7) - identical capabilities with higher output speed at premium 6x pricing.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":30,"output":150,"cached_input":3},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"anthropic/claude-opus-4.7","name":"Anthropic: Claude Opus 4.7","created_by":"anthropic","source":"official","family":"claude-opus","description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.","release_date":"2026-04-16","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":25,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"anthropic/claude-opus-4.8-fast","name":"Anthropic: Claude Opus 4.8 (Fast)","created_by":"anthropic","source":"official","family":"claude-opus","description":"Fast-mode variant of [Opus 4.8](/anthropic/claude-opus-4.8) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.","release_date":"2026-05-27","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Fast-mode variant of [Opus 4.8](/anthropic/claude-opus-4.8) - identical capabilities with higher output speed at 2x pricing relative to regular Opus 4.8.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":10,"output":50,"cached_input":1},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"anthropic/claude-opus-4.8","name":"Anthropic: Claude Opus 4.8","created_by":"anthropic","source":"official","family":"claude-opus","description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family.","release_date":"2026-05-27","context_window":1000000,"max_output_tokens":128000,"model_type":"chat","tagline":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":25,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"anthropic/claude-opus-4","name":"Anthropic: Claude Opus 4","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","description":"Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows.","context_window":200000,"max_output_tokens":32000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":15,"output":75,"cached_input":1.5},"model_type":"chat","release_date":"2025-05-22","tagline":"Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows.","tools":["function_calling"]},{"id":"anthropic/claude-sonnet-4.5","name":"Anthropic: Claude Sonnet 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-sonnet","description":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows.","context_window":1000000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-09-29","tagline":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows.","tools":["function_calling"]},{"id":"anthropic/claude-sonnet-4.6","name":"Anthropic: Claude Sonnet 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-24","family":"claude-sonnet","description":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work.","context_window":1000000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2026-02-17","tagline":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work.","tools":["function_calling"]},{"id":"anthropic/claude-sonnet-4","name":"Anthropic: Claude Sonnet 4","created_by":"anthropic","source":"official","last_updated":"2026-04-12","family":"claude-sonnet","description":"Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability.","context_window":1000000,"max_output_tokens":64000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","release_date":"2025-05-22","tagline":"Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability.","tools":["function_calling"]},{"id":"arcee-ai/coder-large","name":"Arcee AI: Coder Large","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Coder‑Large is a 32 B‑parameter offspring of Qwen 2.5‑Instruct that has been further trained on permissively‑licensed GitHub, CodeSearchNet and synthetic bug‑fix corpora.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":0.8},"release_date":"2025-05-05","tagline":"Coder‑Large is a 32 B‑parameter offspring of Qwen 2.5‑Instruct that has been further trained on permissively‑licensed GitHub, CodeSearchNet and synthetic bug‑fix corpora."},{"id":"arcee-ai/maestro-reasoning","name":"Arcee AI: Maestro Reasoning","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Maestro Reasoning is Arcee's flagship analysis model: a 32 B‑parameter derivative of Qwen 2.5‑32 B tuned with DPO and chain‑of‑thought RL for step‑by‑step logic.","context_window":131072,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.9,"output":3.3},"model_type":"reasoning","release_date":"2025-05-05","tagline":"Maestro Reasoning is Arcee's flagship analysis model: a 32 B‑parameter derivative of Qwen 2.5‑32 B tuned with DPO and chain‑of‑thought RL for step‑by‑step logic."},{"id":"arcee-ai/spotlight","name":"Arcee AI: Spotlight","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Spotlight is a 7‑billion‑parameter vision‑language model derived from Qwen 2.5‑VL and fine‑tuned by Arcee AI for tight image‑text grounding tasks.","context_window":131072,"max_output_tokens":65537,"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.18,"output":0.18},"release_date":"2025-05-05","tagline":"Spotlight is a 7‑billion‑parameter vision‑language model derived from Qwen 2.5‑VL and fine‑tuned by Arcee AI for tight image‑text grounding tasks."},{"id":"arcee-ai/trinity-large-preview:free","name":"Arcee AI: Trinity Large Preview (free)","created_by":"arcee-ai","source":"official","last_updated":"2026-04-10","description":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","context_window":131000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"release_date":"2026-01-27","tagline":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-22"},{"id":"arcee-ai/trinity-large-preview","name":"Arcee AI: Trinity Large Preview","created_by":"arcee-ai","source":"official","description":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","release_date":"2026-01-27","context_window":131000,"tagline":"Trinity-Large-Preview is a frontier-scale open-weight language model from Arcee, built as a 400B-parameter sparse Mixture-of-Experts with 13B active parameters per token using 4-of-256 expert routing.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.15,"output":0.45},"tools":["function_calling"],"last_updated":"2026-04-24"},{"id":"arcee-ai/trinity-large-thinking:free","name":"Arcee AI: Trinity Large Thinking (free)","created_by":"arcee-ai","source":"official","description":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","release_date":"2026-04-01","context_window":262144,"max_output_tokens":80000,"tagline":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-14"},{"id":"arcee-ai/trinity-large-thinking","name":"Arcee AI: Trinity Large Thinking","created_by":"arcee-ai","source":"official","description":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","release_date":"2026-04-01","context_window":262144,"max_output_tokens":80000,"tagline":"Trinity Large Thinking is a powerful open source reasoning model from the team at Arcee AI.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.25,"output":0.8,"cached_input":0.06},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"arcee-ai/trinity-mini:free","name":"Arcee AI: Trinity Mini (free)","created_by":"arcee-ai","source":"official","last_updated":"2026-04-03","description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"release_date":"2025-12-01","tagline":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-10"},{"id":"arcee-ai/trinity-mini","name":"Arcee AI: Trinity Mini","created_by":"arcee-ai","source":"official","last_updated":"2026-03-24","description":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.045,"output":0.15},"release_date":"2025-12-01","tagline":"Trinity Mini is a 26B-parameter (3B active) sparse mixture-of-experts language model featuring 128 experts with 8 active per token.","tools":["function_calling"]},{"id":"arcee-ai/virtuoso-large","name":"Arcee AI: Virtuoso Large","created_by":"arcee-ai","source":"official","last_updated":"2026-04-07","description":"Virtuoso‑Large is Arcee's top‑tier general‑purpose LLM at 72 B parameters, tuned to tackle cross‑domain reasoning, creative writing and enterprise QA.","context_window":131072,"max_output_tokens":64000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.75,"output":1.2},"release_date":"2025-05-05","tagline":"Virtuoso‑Large is Arcee's top‑tier general‑purpose LLM at 72 B parameters, tuned to tackle cross‑domain reasoning, creative writing and enterprise QA.","tools":["function_calling"]},{"id":"baidu/cobuddy:free","name":"Baidu Qianfan: CoBuddy (free)","created_by":"baidu","source":"official","description":"CoBuddy is a code generation model from Baidu, optimized for coding tasks and AI Agent workflows.","release_date":"2026-05-06","context_window":131072,"max_output_tokens":65536,"tagline":"CoBuddy is a code generation model from Baidu, optimized for coding tasks and AI Agent workflows.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-06"},{"id":"baidu/ernie-4.5-21b-a3b-thinking","name":"Baidu: ERNIE 4.5 21B A3B Thinking","created_by":"baidu","source":"official","last_updated":"2026-03-23","description":"ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science, coding, text generation, ","context_window":131072,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.07,"output":0.28},"parameters":21,"active_parameters":3,"release_date":"2025-10-09","tagline":"ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science, coding, text generation, "},{"id":"baidu/ernie-4.5-21b-a3b","name":"Baidu: ERNIE 4.5 21B A3B","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"A sophisticated text-based Mixture-of-Experts (MoE) model featuring 21B total parameters with 3B activated per token, delivering exceptional multimodal understanding and generation through heterogeneous MoE structures and modality-isolated routing.","context_window":131072,"max_output_tokens":8000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.07,"output":0.28},"parameters":21,"active_parameters":3,"release_date":"2025-08-12","tagline":"A sophisticated text-based Mixture-of-Experts (MoE) model featuring 21B total parameters with 3B activated per token, delivering exceptional multimodal understanding and generation through heterogeneo","tools":["function_calling"]},{"id":"baidu/ernie-4.5-300b-a47b","name":"Baidu: ERNIE 4.5 300B A47B ","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"ERNIE-4.5-300B-A47B is a 300B parameter Mixture-of-Experts (MoE) language model developed by Baidu as part of the ERNIE 4.5 series.","context_window":131072,"max_output_tokens":12000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.28,"output":1.1},"parameters":300,"active_parameters":47,"release_date":"2025-06-30","tagline":"ERNIE-4.5-300B-A47B is a 300B parameter Mixture-of-Experts (MoE) language model developed by Baidu as part of the ERNIE 4.5 series."},{"id":"baidu/ernie-4.5-vl-28b-a3b","name":"Baidu: ERNIE 4.5 VL 28B A3B","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"A powerful multimodal Mixture-of-Experts chat model featuring 28B total parameters with 3B activated per token, delivering exceptional text and vision understanding through its innovative heterogeneou","context_window":131072,"max_output_tokens":8000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":0.14,"output":0.56},"parameters":28,"active_parameters":3,"release_date":"2025-08-12","tagline":"A powerful multimodal Mixture-of-Experts chat model featuring 28B total parameters with 3B activated per token, delivering exceptional text and vision understanding through its innovative heterogeneou","tools":["function_calling"]},{"id":"baidu/ernie-4.5-vl-424b-a47b","name":"Baidu: ERNIE 4.5 VL 424B A47B ","created_by":"baidu","source":"official","last_updated":"2026-05-17","description":"ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active per token.","context_window":131072,"max_output_tokens":16000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":0.42,"output":1.25},"parameters":424,"active_parameters":47,"release_date":"2025-06-30","tagline":"ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active per token."},{"id":"baidu/qianfan-ocr-fast:free","name":"Baidu: Qianfan-OCR-Fast (free)","created_by":"baidu","source":"official","description":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","release_date":"2026-04-20","context_window":65536,"max_output_tokens":28672,"tagline":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","capabilities":{"reasoning":true},"modalities":{"input":["image","text"],"output":["text"]},"last_updated":"2026-04-24"},{"id":"baidu/qianfan-ocr-fast","name":"Baidu: Qianfan-OCR-Fast","created_by":"baidu","source":"official","description":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","release_date":"2026-04-20","context_window":65536,"max_output_tokens":28672,"tagline":"Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR.","capabilities":{"reasoning":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.68,"output":2.81},"last_updated":"2026-05-23","status":"deprecated","deprecation_date":"2026-05-28"},{"id":"bytedance-seed/seed-1.6-flash","name":"ByteDance Seed: Seed 1.6 Flash","created_by":"bytedance-seed","source":"official","last_updated":"2026-03-24","description":"Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.075,"output":0.3},"release_date":"2025-12-23","tagline":"Seed 1.6 Flash is an ultra-fast multimodal deep thinking model by ByteDance Seed, supporting both text and visual understanding.","tools":["function_calling"]},{"id":"bytedance-seed/seed-1.6","name":"ByteDance Seed: Seed 1.6","created_by":"bytedance-seed","source":"official","last_updated":"2026-03-24","description":"Seed 1.6 is a general-purpose model released by the ByteDance Seed team.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.25,"output":2},"release_date":"2025-12-23","tagline":"Seed 1.6 is a general-purpose model released by the ByteDance Seed team.","tools":["function_calling"]},{"id":"bytedance-seed/seed-2.0-lite","name":"ByteDance Seed: Seed-2.0-Lite","created_by":"bytedance-seed","source":"official","last_updated":"2026-04-07","description":"Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice f","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.25,"output":2},"release_date":"2026-03-10","tagline":"Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice f","tools":["function_calling"]},{"id":"bytedance-seed/seed-2.0-mini","name":"ByteDance Seed: Seed-2.0-Mini","created_by":"bytedance-seed","source":"official","last_updated":"2026-03-24","description":"Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment.","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"release_date":"2026-02-26","tagline":"Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment.","tools":["function_calling"]},{"id":"bytedance/ui-tars-1.5-7b","name":"ByteDance: UI-TARS 7B ","created_by":"bytedance","source":"official","last_updated":"2026-06-18","description":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games.","context_window":128000,"max_output_tokens":2048,"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":0.1,"output":0.2,"cached_input":0.1},"parameters":7,"release_date":"2025-07-22","tagline":"UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"cognitivecomputations/dolphin-mistral-24b-venice-edition:free","name":"Venice: Uncensored (free)","created_by":"cognitivecomputations","source":"official","last_updated":"2026-03-24","description":"Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"parameters":24,"release_date":"2025-07-09","tagline":"Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai."},{"id":"cohere/command-a","name":"Cohere: Command A","created_by":"cohere","source":"official","last_updated":"2026-03-24","description":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.","context_window":256000,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":2.5,"output":10},"family":"command-a","model_type":"chat","release_date":"2025-03-13","tagline":"Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases."},{"id":"cohere/command-r-08-2024","name":"Cohere: Command R (08-2024)","created_by":"cohere","source":"official","last_updated":"2026-03-24","description":"command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use.","context_window":128000,"max_output_tokens":4000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.6},"family":"command-r","model_type":"chat","release_date":"2024-08-30","tagline":"command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use.","tools":["function_calling"],"license":"cc-by-nc-4.0","open_weight":true},{"id":"cohere/command-r-plus-08-2024","name":"Cohere: Command R+ (08-2024)","created_by":"cohere","source":"official","last_updated":"2026-04-07","description":"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while ","context_window":128000,"max_output_tokens":4000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":2.5,"output":10},"family":"command-r-plus","model_type":"chat","release_date":"2024-08-30","tagline":"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while ","tools":["function_calling"],"license":"cc-by-nc-4.0","open_weight":true},{"id":"cohere/command-r7b-12-2024","name":"Cohere: Command R7B (12-2024)","created_by":"cohere","source":"official","last_updated":"2026-03-24","description":"Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024.","context_window":128000,"max_output_tokens":4000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.038,"output":0.15},"family":"command-r","model_type":"chat","release_date":"2024-12-14","tagline":"Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024.","license":"cc-by-nc-4.0","open_weight":true},{"id":"cohere/north-mini-code:free","name":"Cohere: North Mini Code (free)","created_by":"cohere","source":"official","description":"North Mini Code is Cohere's first agentic coding model and the debut of its North family.","release_date":"2026-06-17","context_window":256000,"max_output_tokens":64000,"tagline":"North Mini Code is Cohere's first agentic coding model and the debut of its North family.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"deepcogito/cogito-v2.1-671b","name":"Deep Cogito: Cogito v2.1 671B","created_by":"deepcogito","source":"official","last_updated":"2026-03-24","description":"Cogito v2.1 671B MoE represents one of the strongest open models globally, matching performance of frontier closed and open models.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.25,"output":1.25},"parameters":671,"release_date":"2025-11-13","tagline":"Cogito v2.1 671B MoE represents one of the strongest open models globally, matching performance of frontier closed and open models.","model_type":"chat"},{"id":"deepseek/deepseek-chat-v3-0324","name":"DeepSeek: DeepSeek V3 0324","created_by":"deepseek","source":"official","last_updated":"2026-04-25","family":"deepseek-chat","description":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","context_window":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.77,"cached_input":0.135},"release_date":"2025-03-24","tagline":"DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.","tools":["function_calling"],"model_type":"chat","max_output_tokens":16384},{"id":"deepseek/deepseek-chat-v3.1","name":"DeepSeek: DeepSeek V3.1","created_by":"deepseek","source":"official","last_updated":"2026-05-14","family":"deepseek-chat","description":"DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.","context_window":163840,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.21,"output":0.79,"cached_input":0.13},"release_date":"2025-08-21","tagline":"DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.","tools":["function_calling"],"model_type":"chat"},{"id":"deepseek/deepseek-chat","name":"DeepSeek: DeepSeek V3","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek-chat","description":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","context_window":131072,"max_output_tokens":16000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.8},"release_date":"2024-12-26","tagline":"DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.","tools":["function_calling"],"model_type":"chat","license":"mit","open_weight":true},{"id":"deepseek/deepseek-r1-0528","name":"DeepSeek: R1 0528","created_by":"deepseek","source":"official","last_updated":"2026-04-25","family":"deepseek-r1","description":"May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","context_window":163840,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.5,"output":2.15,"cached_input":0.35},"model_type":"reasoning","release_date":"2025-05-28","tagline":"May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","tools":["function_calling"]},{"id":"deepseek/deepseek-r1-distill-llama-70b","name":"DeepSeek: R1 Distill Llama 70B","created_by":"deepseek","source":"official","last_updated":"2026-06-18","family":"deepseek-r1","description":"DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1).","context_window":128000,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.8,"output":0.8},"model_type":"reasoning","parameters":70,"release_date":"2025-01-23","tagline":"DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1).","status":"deprecated","deprecation_date":"2026-06-11"},{"id":"deepseek/deepseek-r1-distill-qwen-32b","name":"DeepSeek: R1 Distill Qwen 32B","created_by":"deepseek","source":"official","last_updated":"2026-05-17","family":"deepseek-r1","description":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1).","context_window":128000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.29,"output":0.29},"model_type":"reasoning","parameters":32,"release_date":"2025-01-29","tagline":"DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1)."},{"id":"deepseek/deepseek-r1","name":"DeepSeek: R1","created_by":"deepseek","source":"official","last_updated":"2026-05-17","family":"deepseek-r1","description":"DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","context_window":163840,"max_output_tokens":16000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.7,"output":2.5},"model_type":"reasoning","release_date":"2025-01-20","tagline":"DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens.","tools":["function_calling"]},{"id":"deepseek/deepseek-v3.1-terminus","name":"DeepSeek: DeepSeek V3.1 Terminus","created_by":"deepseek","source":"official","last_updated":"2026-05-06","family":"deepseek","description":"DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language cons","context_window":163840,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.27,"output":0.95,"cached_input":0.13},"release_date":"2025-09-22","tagline":"DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language cons","tools":["function_calling"],"model_type":"chat"},{"id":"deepseek/deepseek-v3.2-exp","name":"DeepSeek: DeepSeek V3.2 Exp","created_by":"deepseek","source":"official","last_updated":"2026-03-24","family":"deepseek","description":"DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.","context_window":163840,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.27,"output":0.41},"release_date":"2025-09-29","tagline":"DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.","tools":["function_calling"],"model_type":"chat"},{"id":"deepseek/deepseek-v3.2-speciale","name":"DeepSeek: DeepSeek V3.2 Speciale","created_by":"deepseek","source":"official","last_updated":"2026-05-07","family":"deepseek","description":"DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance.","context_window":163840,"max_output_tokens":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.287,"output":0.431,"cached_input":0.058},"release_date":"2025-12-01","tagline":"DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized for maximum reasoning and agentic performance.","model_type":"chat"},{"id":"deepseek/deepseek-v3.2","name":"DeepSeek: DeepSeek V3.2","created_by":"deepseek","source":"official","last_updated":"2026-06-29","family":"deepseek","description":"DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.229,"output":0.343,"cached_input":0.023},"release_date":"2025-12-01","tagline":"DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.","tools":["function_calling"],"model_type":"chat","max_output_tokens":64000},{"id":"deepseek/deepseek-v4-flash:free","name":"DeepSeek: DeepSeek V4 Flash (free)","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":384000,"model_type":"chat","tagline":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"deepseek/deepseek-v4-flash","name":"DeepSeek: DeepSeek V4 Flash","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.09,"output":0.18,"cached_input":0.02},"tools":["function_calling"],"last_updated":"2026-06-18","license":"mit","open_weight":true},{"id":"deepseek/deepseek-v4-pro","name":"DeepSeek: DeepSeek V4 Pro","created_by":"deepseek","source":"official","family":"deepseek","description":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.","release_date":"2026-04-24","context_window":1048576,"max_output_tokens":384000,"model_type":"chat","tagline":"DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.435,"output":0.87,"cached_input":0.004},"tools":["function_calling"],"last_updated":"2026-04-26","license":"mit","open_weight":true},{"id":"eleutherai/llemma_7b","name":"EleutherAI: Llemma 7b","created_by":"eleutherai","source":"official","last_updated":"2026-03-23","description":"Llemma 7B is a language model for mathematics.","context_window":4096,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.2},"parameters":7,"release_date":"2025-04-14","tagline":"Llemma 7B is a language model for mathematics."},{"id":"essentialai/rnj-1-instruct","name":"EssentialAI: Rnj 1 Instruct","created_by":"essentialai","source":"official","last_updated":"2026-03-24","description":"Rnj-1 is an 8B-parameter, dense, open-weight model family developed by Essential AI and trained from scratch with a focus on programming, math, and scientific reasoning.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.15,"output":0.15},"release_date":"2025-12-07","tagline":"Rnj-1 is an 8B-parameter, dense, open-weight model family developed by Essential AI and trained from scratch with a focus on programming, math, and scientific reasoning.","tools":["function_calling"],"model_type":"chat"},{"id":"google/gemini-2.0-flash-001","name":"Google: Gemini 2.0 Flash","created_by":"google","source":"official","last_updated":"2026-05-23","family":"gemini-2.0","description":"Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro 1.5](/google/gemini-pro-1.5).","status":"deprecated","deprecation_date":"2026-06-01","context_window":1000000,"max_output_tokens":8192,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.025},"model_type":"chat","release_date":"2025-02-05","tagline":"Gemini Flash 2.0 offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemini Pro","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"google/gemini-2.0-flash-lite-001","name":"Google: Gemini 2.0 Flash Lite","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.0","description":"Gemini 2.0 Flash Lite offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemin","status":"deprecated","deprecation_date":"2026-06-01","context_window":1048576,"max_output_tokens":8192,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.075,"output":0.3},"model_type":"chat","release_date":"2025-02-25","tagline":"Gemini 2.0 Flash Lite offers a significantly faster time to first token (TTFT) compared to [Gemini Flash 1.5](/google/gemini-flash-1.5), while maintaining quality on par with larger models like [Gemin","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-image","name":"Google: Nano Banana (Gemini 2.5 Flash Image)","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","description":"Gemini 2.5 Flash Image, a.k.a.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"batch":true,"vision":true},"pricing":{"input":0.3,"output":2.5,"cached_input":0.03},"model_type":"chat","release_date":"2025-10-07","tagline":"Gemini 2.5 Flash Image, a.k.a.","knowledge_cutoff":"2025-06","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite-preview-09-2025","name":"Google: Gemini 2.5 Flash Lite Preview 09-2025","created_by":"google","source":"official","last_updated":"2026-04-03","family":"gemini-2.5","description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.01},"model_type":"chat","release_date":"2025-09-25","tagline":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite","name":"Google: Gemini 2.5 Flash Lite","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.01},"model_type":"chat","release_date":"2025-07-22","tagline":"Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash","name":"Google: Gemini 2.5 Flash","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["image","text","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.3,"output":2.5,"cached_input":0.03},"model_type":"chat","release_date":"2025-06-17","tagline":"Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-pro-preview-05-06","name":"Google: Gemini 2.5 Pro Preview 05-06","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-05-07","tagline":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-pro-preview","name":"Google: Gemini 2.5 Pro Preview 06-05","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["image","text","audio"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-06-05","tagline":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"]},{"id":"google/gemini-2.5-pro","name":"Google: Gemini 2.5 Pro","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","description":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-06-17","tagline":"Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-flash-preview","name":"Google: Gemini 3 Flash Preview","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemini-3","description":"Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance.","context_window":1048576,"max_output_tokens":65535,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.5,"output":3,"cached_input":0.05},"model_type":"chat","release_date":"2025-12-17","tagline":"Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-image-preview","name":"Google: Nano Banana Pro (Gemini 3 Pro Image Preview)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3","description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","context_window":65536,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2025-11-20","tagline":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-image","name":"Google: Nano Banana Pro (Gemini 3 Pro Image)","created_by":"google","source":"official","family":"gemini-3","description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","release_date":"2026-06-18","context_window":65536,"max_output_tokens":32768,"model_type":"chat","tagline":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["image","text"],"output":["image","text"]},"pricing":{"input":2,"output":12,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-06-18","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-preview","name":"Google: Gemini 3 Pro Preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3","description":"Gemini 3 Pro is Google’s flagship frontier model for high-precision multimodal reasoning, combining strong performance across text, image, video, audio, and code with a 1M-token context window.","status":"deprecated","deprecation_date":"2026-03-26","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2025-11-18","tagline":"Gemini 3 Pro is Google’s flagship frontier model for high-precision multimodal reasoning, combining strong performance across text, image, video, audio, and code with a 1M-token context window.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image-preview","name":"Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-3.1","description":"Gemini 3.1 Flash Image Preview, a.k.a.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","release_date":"2026-02-26","tagline":"Gemini 3.1 Flash Image Preview, a.k.a.","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image","name":"Google: Nano Banana 2 (Gemini 3.1 Flash Image)","created_by":"google","source":"official","family":"gemini-3.1","description":"Gemini 3.1 Flash Image, a.k.a.","release_date":"2026-06-18","context_window":131072,"max_output_tokens":32768,"model_type":"chat","tagline":"Gemini 3.1 Flash Image, a.k.a.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"tool_call":true,"batch":true,"vision":true},"modalities":{"input":["image","text"],"output":["image","text"]},"pricing":{"input":0.5,"output":3},"last_updated":"2026-06-29","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite-preview","name":"Google: Gemini 3.1 Flash Lite Preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3.1","description":"Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["text","image","video","audio"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":0.25,"output":1.5,"cached_input":0.025},"model_type":"chat","release_date":"2026-03-03","tagline":"Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases.","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite","name":"Google: Gemini 3.1 Flash Lite","created_by":"google","source":"official","family":"gemini-3.1","description":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads.","release_date":"2026-05-07","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"pricing":{"input":0.25,"output":1.5,"cached_input":0.025},"tools":["function_calling"],"last_updated":"2026-05-10","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-pro-preview-customtools","name":"Google: Gemini 3.1 Pro Preview Custom Tools","created_by":"google","source":"official","last_updated":"2026-05-17","family":"gemini-3.1","description":"Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party...","context_window":1048756,"max_output_tokens":65536,"modalities":{"input":["text","audio","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2026-02-25","tagline":"Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party or user-defined fu","tools":["function_calling"]},{"id":"google/gemini-3.1-pro-preview","name":"Google: Gemini 3.1 Pro Preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3.1","description":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows.","context_window":1048576,"max_output_tokens":65536,"modalities":{"input":["audio","image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"pricing":{"input":2,"output":12,"cached_input":0.2},"model_type":"chat","release_date":"2026-02-19","tagline":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows","tools":["function_calling"],"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.5-flash","name":"Google: Gemini 3.5 Flash","created_by":"google","source":"official","family":"gemini-3.5","description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed.","release_date":"2026-05-19","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","tagline":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"pricing":{"input":1.5,"output":9,"cached_input":0.15},"tools":["function_calling"],"last_updated":"2026-05-23","knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemma-2-27b-it","name":"Google: Gemma 2 27B","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-2","description":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini).","context_window":8192,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.65,"output":0.65},"model_type":"chat","parameters":27,"release_date":"2024-07-13","tagline":"Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini)."},{"id":"google/gemma-2-9b-it","name":"Google: Gemma 2 9B","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-2","description":"Gemma 2 9B by Google is an advanced, open-source language model that sets a new standard for efficiency and performance in its size class.","context_window":8192,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.09},"model_type":"chat","parameters":9,"release_date":"2024-06-28","tagline":"Gemma 2 9B by Google is an advanced, open-source language model that sets a new standard for efficiency and performance in its size class."},{"id":"google/gemma-3-12b-it:free","name":"Google: Gemma 3 12B (free)","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"model_type":"chat","parameters":12,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs."},{"id":"google/gemma-3-12b-it","name":"Google: Gemma 3 12B","created_by":"google","source":"official","last_updated":"2026-06-10","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true},"pricing":{"input":0.05,"output":0.15},"model_type":"chat","parameters":12,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","max_output_tokens":16384,"tools":["function_calling"]},{"id":"google/gemma-3-27b-it:free","name":"Google: Gemma 3 27B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","parameters":27,"release_date":"2025-03-12","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs."},{"id":"google/gemma-3-27b-it","name":"Google: Gemma 3 27B","created_by":"google","source":"official","last_updated":"2026-04-25","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true},"pricing":{"input":0.08,"output":0.16},"model_type":"chat","parameters":27,"release_date":"2025-03-12","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","tools":["function_calling"]},{"id":"google/gemma-3-4b-it:free","name":"Google: Gemma 3 4B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","parameters":4,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs."},{"id":"google/gemma-3-4b-it","name":"Google: Gemma 3 4B","created_by":"google","source":"official","last_updated":"2026-06-10","family":"gemma-3","description":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.05,"output":0.1},"model_type":"chat","parameters":4,"release_date":"2025-03-13","tagline":"Gemma 3 introduces multimodality, supporting vision-language input and text outputs.","max_output_tokens":16384},{"id":"google/gemma-3n-e2b-it:free","name":"Google: Gemma 3n 2B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3n","description":"Gemma 3n E2B IT is a multimodal, instruction-tuned model developed by Google DeepMind, designed to operate efficiently at an effective parameter size of 2B while leveraging a 6B architecture.","context_window":8192,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","release_date":"2025-07-09","tagline":"Gemma 3n E2B IT is a multimodal, instruction-tuned model developed by Google DeepMind, designed to operate efficiently at an effective parameter size of 2B while leveraging a 6B architecture."},{"id":"google/gemma-3n-e4b-it:free","name":"Google: Gemma 3n 4B (free)","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemma-3n","description":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.","context_window":8192,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"model_type":"chat","release_date":"2025-05-20","tagline":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets."},{"id":"google/gemma-3n-e4b-it","name":"Google: Gemma 3n 4B","created_by":"google","source":"official","last_updated":"2026-06-18","family":"gemma-3n","description":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.06,"output":0.12},"model_type":"chat","release_date":"2025-05-20","tagline":"Gemma 3n E4B-it is optimized for efficient execution on mobile and low-resource devices, such as phones, laptops, and tablets.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"google/gemma-4-26b-a4b-it:free","name":"Google: Gemma 4 26B A4B (free)","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","release_date":"2026-04-03","context_window":262144,"max_output_tokens":32768,"parameters":26,"active_parameters":4,"model_type":"chat","tagline":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-08"},{"id":"google/gemma-4-26b-a4b-it","name":"Google: Gemma 4 26B A4B ","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","release_date":"2026-04-03","context_window":262144,"max_output_tokens":16384,"parameters":26,"active_parameters":4,"model_type":"chat","tagline":"Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.06,"output":0.33,"cached_input":0.04},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"google/gemma-4-31b-it:free","name":"Google: Gemma 4 31B (free)","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","release_date":"2026-04-02","context_window":262144,"max_output_tokens":8192,"parameters":31,"model_type":"chat","tagline":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"google/gemma-4-31b-it","name":"Google: Gemma 4 31B","created_by":"google","source":"official","family":"gemma-4","description":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","release_date":"2026-04-02","context_window":262144,"max_output_tokens":262144,"parameters":31,"model_type":"chat","tagline":"Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"tool_call":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.12,"output":0.35,"cached_input":0.09},"last_updated":"2026-06-18","tools":["function_calling"]},{"id":"google/lyria-3-clip-preview","name":"Google: Lyria 3 Clip Preview","created_by":"google","source":"official","description":"30 second duration clips are priced at $0.04 per clip.","release_date":"2026-03-30","context_window":1048576,"max_output_tokens":65536,"model_type":"tts","tagline":"30 second duration clips are priced at $0.04 per clip.","capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text","audio"]},"last_updated":"2026-04-01","license":"proprietary","open_weight":false},{"id":"google/lyria-3-pro-preview","name":"Google: Lyria 3 Pro Preview","created_by":"google","source":"official","description":"Full-length songs are priced at $0.08 per song.","release_date":"2026-03-30","context_window":1048576,"max_output_tokens":65536,"model_type":"tts","tagline":"Full-length songs are priced at $0.08 per song.","capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"tool_call":true,"reasoning":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text","audio"]},"last_updated":"2026-04-01","license":"proprietary","open_weight":false},{"id":"gryphe/mythomax-l2-13b","name":"MythoMax 13B","created_by":"gryphe","source":"official","last_updated":"2026-03-24","description":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay.","context_window":4096,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.06,"output":0.06},"parameters":13,"release_date":"2023-07-02","tagline":"One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay."},{"id":"ibm-granite/granite-4.0-h-micro","name":"IBM: Granite 4.0 Micro","created_by":"ibm-granite","source":"official","last_updated":"2026-06-18","description":"Granite-4.0-H-Micro is a 3B parameter from the Granite 4 family of models.","context_window":131000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.017,"output":0.112},"release_date":"2025-10-20","tagline":"Granite-4.0-H-Micro is a 3B parameter from the Granite 4 family of models.","model_type":"chat","max_output_tokens":131000,"capabilities":{"structured_output":true,"json_mode":true}},{"id":"ibm-granite/granite-4.1-8b","name":"IBM: Granite 4.1 8B","created_by":"ibm-granite","source":"official","description":"Granite 4.1 8B is a dense, decoder-only 8-billion-parameter language model from IBM, part of the Granite 4.1 family.","release_date":"2026-04-30","context_window":131072,"max_output_tokens":131072,"parameters":8,"model_type":"chat","tagline":"Granite 4.1 8B is a dense, decoder-only 8-billion-parameter language model from IBM, part of the Granite 4.1 family.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.1,"cached_input":0.05},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"inception/mercury-2","name":"Inception: Mercury 2","created_by":"inception","source":"official","last_updated":"2026-03-24","description":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM).","context_window":128000,"max_output_tokens":50000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":0.75,"cached_input":0.025},"release_date":"2026-03-04","tagline":"Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM).","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"inception/mercury-coder","name":"Inception: Mercury Coder","created_by":"inception","source":"official","last_updated":"2026-04-03","description":"Mercury Coder is the first diffusion large language model (dLLM).","context_window":128000,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":0.75,"cached_input":0.025},"release_date":"2025-04-30","tagline":"Mercury Coder is the first diffusion large language model (dLLM).","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-15","license":"proprietary","open_weight":false},{"id":"inception/mercury","name":"Inception: Mercury","created_by":"inception","source":"official","last_updated":"2026-03-24","description":"Mercury is the first diffusion large language model (dLLM).","status":"deprecated","deprecation_date":"2026-04-15","context_window":128000,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true},"pricing":{"input":0.25,"output":0.75,"cached_input":0.025},"release_date":"2025-06-26","tagline":"Mercury is the first diffusion large language model (dLLM).","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"inclusionai/ling-2.6-1t:free","name":"inclusionAI: Ling-2.6-1T (free)","created_by":"inclusionai","source":"official","description":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","status":"deprecated","release_date":"2026-04-23","deprecation_date":"2026-05-07","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"inclusionai/ling-2.6-1t","name":"inclusionAI: Ling-2.6-1T","created_by":"inclusionai","source":"official","description":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","release_date":"2026-04-23","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-1T is an instant (instruct) model from inclusionAI and the company’s trillion-parameter flagship, designed for real-world agents that require fast execution and high efficiency at scale.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.075,"output":0.625,"cached_input":0.015},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"inclusionai/ling-2.6-flash:free","name":"inclusionAI: Ling-2.6-flash (free)","created_by":"inclusionai","source":"official","description":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","release_date":"2026-04-21","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-24","status":"deprecated","deprecation_date":"2026-04-29"},{"id":"inclusionai/ling-2.6-flash","name":"inclusionAI: Ling-2.6-flash","created_by":"inclusionai","source":"official","description":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","release_date":"2026-04-21","context_window":262144,"max_output_tokens":32768,"tagline":"Ling-2.6-flash is an instant (instruct) model from inclusionAI with 104B total parameters and 7.4B active parameters, designed for real-world agents that require fast responses, strong execution, and ","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.01,"output":0.03,"cached_input":0.002},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"inclusionai/ring-2.6-1t:free","name":"inclusionAI: Ring-2.6-1T (free)","created_by":"inclusionai","source":"official","description":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","release_date":"2026-05-08","context_window":262144,"max_output_tokens":65536,"tagline":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-10"},{"id":"inclusionai/ring-2.6-1t","name":"inclusionAI: Ring-2.6-1T","created_by":"inclusionai","source":"official","description":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","release_date":"2026-05-08","context_window":262144,"max_output_tokens":65536,"tagline":"Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.075,"output":0.625,"cached_input":0.015},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"inflection/inflection-3-pi","name":"Inflection: Inflection 3 Pi","created_by":"inflection","source":"official","last_updated":"2026-03-23","description":"Inflection 3 Pi powers Inflection's [Pi](https://pi.ai) chatbot, including backstory, emotional intelligence, productivity, and safety.","context_window":8000,"max_output_tokens":1024,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.5,"output":10},"release_date":"2024-10-11","tagline":"Inflection 3 Pi powers Inflection's [Pi](https://pi.ai) chatbot, including backstory, emotional intelligence, productivity, and safety."},{"id":"inflection/inflection-3-productivity","name":"Inflection: Inflection 3 Productivity","created_by":"inflection","source":"official","last_updated":"2026-03-23","description":"Inflection 3 Productivity is optimized for following instructions.","context_window":8000,"max_output_tokens":1024,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.5,"output":10},"release_date":"2024-10-11","tagline":"Inflection 3 Productivity is optimized for following instructions."},{"id":"kwaipilot/kat-coder-pro-v2","name":"Kwaipilot: KAT-Coder-Pro V2","created_by":"kwaipilot","source":"official","description":"KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and SaaS integration.","release_date":"2026-03-27","context_window":256000,"max_output_tokens":80000,"tagline":"KAT-Coder-Pro V2 is the latest high-performance model in KwaiKAT’s KAT-Coder series, designed for complex enterprise-grade software engineering and SaaS integration.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"kwaipilot/kat-coder-pro","name":"Kwaipilot: KAT-Coder-Pro V1","created_by":"kwaipilot","source":"official","last_updated":"2026-03-24","description":"KAT-Coder-Pro V1 is KwaiKAT's most advanced agentic coding model in the KAT-Coder series.","context_window":256000,"max_output_tokens":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.207,"output":0.828,"cached_input":0.041},"release_date":"2025-11-10","tagline":"KAT-Coder-Pro V1 is KwaiKAT's most advanced agentic coding model in the KAT-Coder series.","tools":["function_calling"]},{"id":"liquid/lfm-2-24b-a2b","name":"LiquidAI: LFM2-24B-A2B","created_by":"liquid","source":"official","last_updated":"2026-06-18","description":"LFM2-24B-A2B is the largest model in the LFM2 family of hybrid architectures designed for efficient on-device deployment.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.03,"output":0.12},"parameters":24,"active_parameters":2,"release_date":"2026-02-25","tagline":"LFM2-24B-A2B is the largest model in the LFM2 family of hybrid architectures designed for efficient on-device deployment.","model_type":"chat","capabilities":{"structured_output":true,"json_mode":true}},{"id":"liquid/lfm-2.2-6b","name":"LiquidAI: LFM2-2.6B","created_by":"liquid","source":"official","last_updated":"2026-03-23","description":"LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.01,"output":0.02},"parameters":6,"release_date":"2025-10-20","tagline":"LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment.","model_type":"chat"},{"id":"liquid/lfm-2.5-1.2b-instruct:free","name":"LiquidAI: LFM2.5-1.2B-Instruct (free)","created_by":"liquid","source":"official","last_updated":"2026-06-18","description":"LFM2.5-1.2B-Instruct is a compact, high-performance instruction-tuned model built for fast on-device AI.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"parameters":1.2,"release_date":"2026-01-20","tagline":"LFM2.5-1.2B-Instruct is a compact, high-performance instruction-tuned model built for fast on-device AI.","model_type":"chat","capabilities":{"structured_output":true,"json_mode":true}},{"id":"liquid/lfm-2.5-1.2b-thinking:free","name":"LiquidAI: LFM2.5-1.2B-Thinking (free)","created_by":"liquid","source":"official","last_updated":"2026-06-18","description":"LFM2.5-1.2B-Thinking is a lightweight reasoning-focused model optimized for agentic tasks, data extraction, and RAG—while still running comfortably on edge devices.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true,"tool_call":true,"structured_output":true,"json_mode":true},"parameters":1.2,"release_date":"2026-01-20","tagline":"LFM2.5-1.2B-Thinking is a lightweight reasoning-focused model optimized for agentic tasks, data extraction, and RAG—while still running comfortably on edge devices.","model_type":"chat","tools":["function_calling"]},{"id":"liquid/lfm2-8b-a1b","name":"LiquidAI: LFM2-8B-A1B","created_by":"liquid","source":"official","last_updated":"2026-03-23","description":"LFM2-8B-A1B is an efficient on-device Mixture-of-Experts (MoE) model from Liquid AI’s LFM2 family, built for fast, high-quality inference on edge hardware.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.01,"output":0.02},"parameters":8,"active_parameters":1,"release_date":"2025-10-20","tagline":"LFM2-8B-A1B is an efficient on-device Mixture-of-Experts (MoE) model from Liquid AI’s LFM2 family, built for fast, high-quality inference on edge hardware.","model_type":"chat"},{"id":"mancer/weaver","name":"Mancer: Weaver (alpha)","created_by":"mancer","source":"official","last_updated":"2026-03-24","description":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory.","context_window":8000,"max_output_tokens":2000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.75,"output":1},"release_date":"2023-08-02","tagline":"An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory."},{"id":"meituan/longcat-flash-chat","name":"Meituan: LongCat Flash Chat","created_by":"meituan","source":"official","last_updated":"2026-03-24","description":"LongCat-Flash-Chat is a large-scale Mixture-of-Experts (MoE) model with 560B total parameters, of which 18.6B–31.3B (≈27B on average) are dynamically activated per input.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.8,"cached_input":0.2},"release_date":"2025-09-09","tagline":"LongCat-Flash-Chat is a large-scale Mixture-of-Experts (MoE) model with 560B total parameters, of which 18.6B–31.3B (≈27B on average) are dynamically activated per input.","tools":["function_calling"]},{"id":"meta-llama/llama-3-70b-instruct","name":"Meta: Llama 3 70B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","context_window":8192,"max_output_tokens":8000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.51,"output":0.74},"model_type":"chat","parameters":70,"release_date":"2024-04-18","tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","status":"deprecated","deprecation_date":"2026-06-19"},{"id":"meta-llama/llama-3-8b-instruct","name":"Meta: Llama 3 8B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3","description":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","context_window":8192,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.14,"output":0.14},"model_type":"chat","parameters":8,"release_date":"2024-04-18","tagline":"Meta's latest class of model (Llama 3) launched with a variety of sizes & flavors.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"meta-llama/llama-3.1-405b","name":"Meta: Llama 3.1 405B (base)","created_by":"meta-llama","source":"official","last_updated":"2026-03-23","family":"llama-3.1","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":4,"output":4},"model_type":"chat","parameters":405,"release_date":"2024-08-02","tagline":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors."},{"id":"meta-llama/llama-3.1-70b-instruct","name":"Meta: Llama 3.1 70B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-04-25","family":"llama-3.1","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":0.4},"model_type":"chat","parameters":70,"release_date":"2024-07-23","tagline":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","tools":["function_calling"],"max_output_tokens":16384},{"id":"meta-llama/llama-3.1-8b-instruct","name":"Meta: Llama 3.1 8B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3.1","description":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.02,"output":0.03},"model_type":"chat","parameters":8,"release_date":"2024-07-23","tagline":"Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.","tools":["function_calling"]},{"id":"meta-llama/llama-3.2-11b-vision-instruct","name":"Meta: Llama 3.2 11B Vision Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-3.2","description":"Llama 3.2 11B Vision is a multimodal model with 11 billion parameters, designed to handle tasks combining visual and textual data.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.345,"output":0.345},"model_type":"chat","parameters":11,"release_date":"2024-09-25","tagline":"Llama 3.2 11B Vision is a multimodal model with 11 billion parameters, designed to handle tasks combining visual and textual data."},{"id":"meta-llama/llama-3.2-1b-instruct","name":"Meta: Llama 3.2 1B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-05-23","family":"llama-3.2","description":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.027,"output":0.201},"model_type":"chat","parameters":1,"release_date":"2024-09-25","tagline":"Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis.","max_output_tokens":60000},{"id":"meta-llama/llama-3.2-3b-instruct:free","name":"Meta: Llama 3.2 3B Instruct (free)","created_by":"meta-llama","source":"official","last_updated":"2026-03-23","family":"llama-3.2","description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","parameters":3,"release_date":"2024-09-25","tagline":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization."},{"id":"meta-llama/llama-3.2-3b-instruct","name":"Meta: Llama 3.2 3B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-05-23","family":"llama-3.2","description":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.051,"output":0.335},"model_type":"chat","parameters":3,"release_date":"2024-09-25","tagline":"Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization.","max_output_tokens":80000},{"id":"meta-llama/llama-3.3-70b-instruct:free","name":"Meta: Llama 3.3 70B Instruct (free)","created_by":"meta-llama","source":"official","last_updated":"2026-05-17","family":"llama-3.3","description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"model_type":"chat","parameters":70,"release_date":"2024-12-06","tagline":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","tools":["function_calling"]},{"id":"meta-llama/llama-3.3-70b-instruct","name":"Meta: Llama 3.3 70B Instruct","created_by":"meta-llama","source":"official","last_updated":"2026-04-22","family":"llama-3.3","description":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.32},"model_type":"chat","parameters":70,"release_date":"2024-12-06","tagline":"The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).","tools":["function_calling"]},{"id":"meta-llama/llama-4-maverick","name":"Meta: Llama 4 Maverick","created_by":"meta-llama","source":"official","last_updated":"2026-04-07","family":"llama-4","description":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forw","context_window":1048576,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","release_date":"2025-04-05","tagline":"Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forw","tools":["function_calling"]},{"id":"meta-llama/llama-4-scout","name":"Meta: Llama 4 Scout","created_by":"meta-llama","source":"official","last_updated":"2026-06-10","family":"llama-4","description":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","context_window":10000000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","release_date":"2025-04-05","tagline":"Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B.","tools":["function_calling"]},{"id":"meta-llama/llama-guard-3-8b","name":"Llama Guard 3 8B","created_by":"meta-llama","source":"official","last_updated":"2026-05-23","family":"llama-guard","description":"Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.484,"output":0.03},"model_type":"moderation","parameters":8,"release_date":"2025-02-12","tagline":"Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification.","max_output_tokens":131072},{"id":"meta-llama/llama-guard-4-12b:free","name":"Meta: Llama Guard 4 12B (free)","created_by":"meta-llama","source":"official","family":"llama-guard","description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","release_date":"2025-04-30","context_window":163840,"max_output_tokens":65000,"model_type":"chat","tagline":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","modalities":{"input":["image","text"],"output":["text"]},"last_updated":"2026-04-16"},{"id":"meta-llama/llama-guard-4-12b","name":"Meta: Llama Guard 4 12B","created_by":"meta-llama","source":"official","last_updated":"2026-04-25","family":"llama-guard","description":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","context_window":163840,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.18,"output":0.18},"model_type":"moderation","parameters":12,"release_date":"2025-04-30","tagline":"Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification.","max_output_tokens":16384},{"id":"microsoft/phi-4-mini-instruct","name":"Microsoft: Phi 4 Mini Instruct","created_by":"microsoft","source":"official","description":"Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data.","release_date":"2025-10-17","context_window":131072,"max_output_tokens":128000,"model_type":"chat","tagline":"Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data.","capabilities":{"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.08,"output":0.35,"cached_input":0.08},"last_updated":"2026-05-17"},{"id":"microsoft/phi-4","name":"Microsoft: Phi 4","created_by":"microsoft","source":"official","last_updated":"2026-06-29","description":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed.","context_window":16384,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.07,"output":0.14},"release_date":"2025-01-10","tagline":"[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed.","model_type":"chat"},{"id":"microsoft/wizardlm-2-8x22b","name":"WizardLM-2 8x22B","created_by":"microsoft","source":"official","last_updated":"2026-05-17","description":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","context_window":65536,"max_output_tokens":8000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.62,"output":0.62},"release_date":"2024-04-16","tagline":"WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"minimax/minimax-01","name":"MiniMax: MiniMax-01","created_by":"minimax","source":"official","last_updated":"2026-03-23","description":"MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding.","context_window":1000192,"max_output_tokens":1000192,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":1.1},"model_type":"chat","family":"minimax","release_date":"2025-01-15","tagline":"MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding."},{"id":"minimax/minimax-m1","name":"MiniMax: MiniMax M1","created_by":"minimax","source":"official","last_updated":"2026-03-23","description":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference.","context_window":1000000,"max_output_tokens":40000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"pricing":{"input":0.4,"output":2.2},"model_type":"chat","family":"minimax","release_date":"2025-06-17","tagline":"MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference.","tools":["function_calling"]},{"id":"minimax/minimax-m2-her","name":"MiniMax: MiniMax M2-her","created_by":"minimax","source":"official","last_updated":"2026-03-23","description":"MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations.","context_window":65536,"max_output_tokens":2048,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":1.2,"cached_input":0.03},"model_type":"chat","family":"minimax","release_date":"2026-01-23","tagline":"MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations."},{"id":"minimax/minimax-m2.1","name":"MiniMax: MiniMax M2.1","created_by":"minimax","source":"official","last_updated":"2026-05-17","description":"MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development.","context_window":204800,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.29,"output":0.95,"cached_input":0.03},"family":"minimax","model_type":"chat","release_date":"2025-12-23","tagline":"MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development.","tools":["function_calling"],"max_output_tokens":196608,"license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5:free","name":"MiniMax: MiniMax M2.5 (free)","created_by":"minimax","source":"official","last_updated":"2026-05-17","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","context_window":204800,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"family":"minimax","model_type":"chat","release_date":"2026-02-12","tagline":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","tools":["function_calling"]},{"id":"minimax/minimax-m2.5","name":"MiniMax: MiniMax M2.5","created_by":"minimax","source":"official","last_updated":"2026-06-29","description":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","context_window":204800,"max_output_tokens":196608,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.12,"output":0.48,"cached_input":0.05},"family":"minimax","model_type":"chat","release_date":"2026-02-12","tagline":"MiniMax-M2.5 is a SOTA large language model designed for real-world productivity.","tools":["function_calling"],"license":"mit","open_weight":true},{"id":"minimax/minimax-m2.7","name":"MiniMax: MiniMax M2.7","created_by":"minimax","source":"official","last_updated":"2026-06-29","description":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.","context_window":204800,"max_output_tokens":196608,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":0.18,"output":0.72,"cached_input":0.05},"family":"minimax","model_type":"chat","release_date":"2026-03-18","tagline":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2","name":"MiniMax: MiniMax M2","created_by":"minimax","source":"official","last_updated":"2026-05-17","description":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","context_window":204800,"max_output_tokens":196608,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":0.255,"output":1,"cached_input":0.03},"model_type":"chat","family":"minimax","release_date":"2025-10-23","tagline":"MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"minimax/minimax-m3","name":"MiniMax: MiniMax M3","created_by":"minimax","source":"official","family":"minimax","description":"MiniMax-M3 is a multimodal foundation model from MiniMax.","release_date":"2026-05-31","context_window":1048576,"max_output_tokens":512000,"model_type":"chat","tagline":"MiniMax-M3 is a multimodal foundation model from MiniMax.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.3,"output":1.2,"cached_input":0.06},"tools":["function_calling"],"last_updated":"2026-06-02","license":"proprietary","open_weight":false},{"id":"mistralai/codestral-2508","name":"Mistral: Codestral 2508","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"codestral","description":"Mistral's cutting-edge language model for coding released end of July 2025.","context_window":256000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.3,"output":0.9,"cached_input":0.03},"model_type":"code","release_date":"2025-08-01","alias":"mistralai/codestral","tagline":"Mistral's cutting-edge language model for coding released end of July 2025.","tools":["function_calling"],"license":"mnpl","open_weight":true},{"id":"mistralai/codestral","name":"Mistral: Codestral 2508","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"codestral","description":"Mistral's cutting-edge language model for coding released end of July 2025.","release_date":"2025-08-01","context_window":256000,"model_type":"code","tagline":"Mistral's cutting-edge language model for coding released end of July 2025.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.9,"cached_input":0.03},"tools":["function_calling"],"snapshots":["mistralai/codestral-2508"],"license":"mnpl","open_weight":true},{"id":"mistralai/devstral-2512","name":"Mistral: Devstral 2 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"code","release_date":"2025-12-09","alias":"mistralai/devstral","tagline":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","tools":["function_calling"]},{"id":"mistralai/devstral-medium","name":"Mistral: Devstral Medium","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"chat","release_date":"2025-07-10","tagline":"Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"mistralai/devstral-small","name":"Mistral: Devstral Small 1.1","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"model_type":"chat","release_date":"2025-07-10","tagline":"Devstral Small 1.1 is a 24B parameter open-weight language model for software engineering agents, developed by Mistral AI in collaboration with All Hands AI.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/devstral","name":"Mistral: Devstral 2 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"devstral","description":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","release_date":"2025-12-09","context_window":262144,"model_type":"code","tagline":"Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"tools":["function_calling"],"snapshots":["mistralai/devstral-2512"]},{"id":"mistralai/ministral-14b-2512","name":"Mistral: Ministral 3 14B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.2,"cached_input":0.02},"model_type":"chat","parameters":14,"release_date":"2025-12-02","alias":"mistralai/ministral-14b","tagline":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-14b","name":"Mistral: Ministral 3 14B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","release_date":"2025-12-02","context_window":262144,"parameters":14,"model_type":"chat","tagline":"The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2,"cached_input":0.02},"tools":["function_calling"],"snapshots":["mistralai/ministral-14b-2512"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-3b-2512","name":"Mistral: Ministral 3 3B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.1,"cached_input":0.01},"model_type":"chat","parameters":3,"release_date":"2025-12-02","alias":"mistralai/ministral-3b","tagline":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-3b","name":"Mistral: Ministral 3 3B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","release_date":"2025-12-02","context_window":131072,"parameters":3,"model_type":"chat","tagline":"The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1,"cached_input":0.01},"tools":["function_calling"],"snapshots":["mistralai/ministral-3b-2512"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-8b-2512","name":"Mistral: Ministral 3 8B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15,"cached_input":0.015},"model_type":"chat","parameters":8,"release_date":"2025-12-02","alias":"mistralai/ministral-8b","tagline":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/ministral-8b","name":"Mistral: Ministral 3 8B 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"ministral","description":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","release_date":"2025-12-02","context_window":262144,"parameters":8,"model_type":"chat","tagline":"A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.15,"cached_input":0.015},"tools":["function_calling"],"snapshots":["mistralai/ministral-8b-2512"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-7b-instruct-v0.1","name":"Mistral: Mistral 7B Instruct v0.1","created_by":"mistral","source":"official","last_updated":"2026-05-23","description":"A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length.","context_window":4096,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.11,"output":0.19},"model_type":"chat","family":"mistral-7b","parameters":7,"release_date":"2023-09-28","tagline":"A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length.","status":"deprecated","deprecation_date":"2026-05-30","max_output_tokens":2824},{"id":"mistralai/mistral-large-2407","name":"Mistral Large 2407","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-large","description":"This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407).","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-11-19","alias":"mistralai/mistral-large","tagline":"This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407).","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-large-2411","name":"Mistral Large 2411","created_by":"mistral","source":"official","last_updated":"2026-04-07","family":"mistral-large","description":"Mistral Large 2 2411 is an update of [Mistral Large 2](/mistralai/mistral-large) released together with [Pixtral Large 2411](/mistralai/pixtral-large-2411) It provides a significant upgrade on the pre","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-11-19","alias":"mistralai/mistral-large","tagline":"Mistral Large 2 2411 is an update of [Mistral Large 2](/mistralai/mistral-large) released together with [Pixtral Large 2411](/mistralai/pixtral-large-2411)\n\nIt provides a significant upgrade on the pr","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-large-2512","name":"Mistral: Mistral Large 3 2512","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-large","description":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.5,"output":1.5,"cached_input":0.05},"model_type":"chat","release_date":"2025-12-01","alias":"mistralai/mistral-large","tagline":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-large","name":"Mistral Large","created_by":"mistral","source":"official","last_updated":"2026-03-26","family":"mistral-large","description":"This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`).","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-02-26","tagline":"Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.","tools":["function_calling"],"snapshots":["mistralai/mistral-large-2512","mistralai/mistral-large-2411","mistralai/mistral-large-2407"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-medium-3-5","name":"Mistral: Mistral Medium 3.5","created_by":"mistral","source":"official","family":"mistral-medium","description":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI.","release_date":"2026-04-30","context_window":262144,"model_type":"chat","tagline":"Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.5,"output":7.5},"tools":["function_calling"],"last_updated":"2026-05-06","license":"proprietary","open_weight":false},{"id":"mistralai/mistral-medium-3.1","name":"Mistral: Mistral Medium 3.1","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-medium","description":"Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"chat","release_date":"2025-08-13","tagline":"Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced opera","tools":["function_calling"]},{"id":"mistralai/mistral-medium-3","name":"Mistral: Mistral Medium 3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-medium","description":"Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost.","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":2,"cached_input":0.04},"model_type":"chat","release_date":"2025-05-07","tagline":"Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost.","tools":["function_calling"]},{"id":"mistralai/mistral-nemo","name":"Mistral: Mistral Nemo","created_by":"mistral","source":"official","last_updated":"2026-05-23","description":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.02,"output":0.03},"model_type":"chat","family":"mistral-nemo","release_date":"2024-07-19","tagline":"A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-saba","name":"Mistral: Saba","created_by":"mistral","source":"official","last_updated":"2026-03-24","description":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.6,"cached_input":0.02},"model_type":"chat","family":"mistral-saba","release_date":"2025-02-17","tagline":"Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-small-24b-instruct-2501","name":"Mistral: Mistral Small 3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","context_window":32768,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.05,"output":0.08},"model_type":"chat","parameters":24,"release_date":"2025-01-30","alias":"mistralai/mistral-small-24b-instruct","tagline":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","tools":["function_calling"]},{"id":"mistralai/mistral-small-24b-instruct","name":"Mistral: Mistral Small 3","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","release_date":"2025-01-30","context_window":32768,"max_output_tokens":16384,"parameters":24,"model_type":"chat","tagline":"Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.05,"output":0.08},"tools":["function_calling"],"snapshots":["mistralai/mistral-small-24b-instruct-2501"]},{"id":"mistralai/mistral-small-2603","name":"Mistral: Mistral Small 4","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.015},"model_type":"chat","release_date":"2026-03-16","alias":"mistralai/mistral-small","tagline":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-small-3.1-24b-instruct:free","name":"Mistral: Mistral Small 3.1 24B (free)","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities.","status":"deprecated","deprecation_date":"2026-03-29","context_window":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"model_type":"chat","parameters":24,"release_date":"2025-03-17","tagline":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities.","tools":["function_calling"]},{"id":"mistralai/mistral-small-3.1-24b-instruct","name":"Mistral: Mistral Small 3.1 24B","created_by":"mistral","source":"official","last_updated":"2026-05-23","family":"mistral-small","description":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities.","context_window":128000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.351,"output":0.555,"cached_input":0.015},"model_type":"chat","parameters":24,"release_date":"2025-03-17","tagline":"Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities."},{"id":"mistralai/mistral-small-3.2-24b-instruct","name":"Mistral: Mistral Small 3.2 24B","created_by":"mistral","source":"official","last_updated":"2026-04-25","family":"mistral-small","description":"Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling.","context_window":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.075,"output":0.2},"model_type":"chat","parameters":24,"release_date":"2025-06-20","tagline":"Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling.","tools":["function_calling"],"max_output_tokens":16384},{"id":"mistralai/mistral-small-creative","name":"Mistral: Mistral Small Creative","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","description":"Mistral Small Creative is an experimental small model designed for creative writing, narrative generation, roleplay and character-driven dialogue, general-purpose instruction following, and conversati","status":"deprecated","deprecation_date":"2026-04-30","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"model_type":"chat","release_date":"2025-12-16","tagline":"Mistral Small Creative is an experimental small model designed for creative writing, narrative generation, roleplay and character-driven dialogue, general-purpose instruction following, and conversati","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mistral-small","name":"Mistral: Mistral Small 4","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mistral-small","description":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","release_date":"2026-03-16","context_window":262144,"model_type":"chat","tagline":"Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.15,"output":0.6,"cached_input":0.015},"tools":["function_calling"],"snapshots":["mistralai/mistral-small-2603"],"license":"apache-2.0","open_weight":true},{"id":"mistralai/mixtral-8x22b-instruct","name":"Mistral: Mixtral 8x22B Instruct","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"mixtral","description":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b).","context_window":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-04-17","tagline":"Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b).","tools":["function_calling"]},{"id":"mistralai/mixtral-8x7b-instruct","name":"Mistral: Mixtral 8x7B Instruct","created_by":"mistral","source":"official","last_updated":"2026-05-07","family":"mixtral","description":"Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use.","context_window":32768,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.54,"output":0.54},"model_type":"chat","release_date":"2023-12-10","tagline":"Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"mistralai/pixtral-12b","name":"Mistral: Pixtral 12B","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"pixtral","description":"The first multi-modal, text+image-to-text model from Mistral AI.","context_window":32768,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","parameters":12,"release_date":"2024-09-10","tagline":"The first multi-modal, text+image-to-text model from Mistral AI.","license":"apache-2.0","open_weight":true,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true}},{"id":"mistralai/pixtral-large-2411","name":"Mistral: Pixtral Large 2411","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"pixtral","description":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","context_window":131072,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2024-11-19","alias":"mistralai/pixtral-large","tagline":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","tools":["function_calling"],"license":"mnpl","open_weight":true},{"id":"mistralai/pixtral-large","name":"Mistral: Pixtral Large 2411","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"pixtral","description":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","release_date":"2024-11-19","context_window":131072,"model_type":"chat","tagline":"Pixtral Large is a 124B parameter, open-weight, multimodal model built on top of [Mistral Large 2](/mistralai/mistral-large-2411).","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":6,"cached_input":0.2},"tools":["function_calling"],"snapshots":["mistralai/pixtral-large-2411"],"license":"mnpl","open_weight":true},{"id":"mistralai/voxtral-small-24b-2507","name":"Mistral: Voxtral Small 24B 2507","created_by":"mistral","source":"official","last_updated":"2026-03-24","description":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","context_window":32000,"modalities":{"input":["text","audio"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"model_type":"audio","family":"voxtral","parameters":24,"release_date":"2025-10-30","alias":"mistralai/voxtral-small-24b","tagline":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","tools":["function_calling"]},{"id":"mistralai/voxtral-small-24b","name":"Mistral: Voxtral Small 24B 2507","created_by":"mistral","source":"official","last_updated":"2026-03-24","family":"voxtral","description":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","release_date":"2025-10-30","context_window":32000,"parameters":24,"model_type":"audio","tagline":"Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text","audio"],"output":["text"]},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"tools":["function_calling"],"snapshots":["mistralai/voxtral-small-24b-2507"]},{"id":"moonshotai/kimi-k2-0905","name":"MoonshotAI: Kimi K2 0905","created_by":"moonshotai","source":"official","last_updated":"2026-06-29","description":"Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2).","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.6,"output":2.5,"cached_input":0.15},"family":"kimi-k2","model_type":"chat","release_date":"2025-09-04","tagline":"Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2).","tools":["function_calling"],"max_output_tokens":100352,"status":"deprecated","deprecation_date":"2026-05-14"},{"id":"moonshotai/kimi-k2-thinking","name":"MoonshotAI: Kimi K2 Thinking","created_by":"moonshotai","source":"official","last_updated":"2026-06-29","description":"Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":2.5,"cached_input":0.6},"family":"kimi-k2","model_type":"chat","release_date":"2025-11-06","tagline":"Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning.","tools":["function_calling"],"max_output_tokens":262144},{"id":"moonshotai/kimi-k2.5","name":"MoonshotAI: Kimi K2.5","created_by":"moonshotai","source":"official","last_updated":"2026-06-18","description":"Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.375,"output":2.025,"cached_input":0.09},"family":"kimi-k2.5","model_type":"chat","release_date":"2026-01-27","tagline":"Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm.","tools":["function_calling"]},{"id":"moonshotai/kimi-k2.6:free","name":"MoonshotAI: Kimi K2.6 (free)","created_by":"moonshotai","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","release_date":"2026-04-20","context_window":262144,"model_type":"chat","tagline":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"moonshotai/kimi-k2.6","name":"MoonshotAI: Kimi K2.6","created_by":"moonshotai","source":"official","family":"kimi-k2.6","description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","release_date":"2026-04-20","context_window":262144,"max_output_tokens":262144,"model_type":"chat","tagline":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.55,"output":3.2,"cached_input":0.11},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"moonshotai/kimi-k2.7-code","name":"MoonshotAI: Kimi K2.7 Code","created_by":"moonshotai","source":"official","family":"kimi-k2.7","description":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.","release_date":"2026-06-12","context_window":262144,"max_output_tokens":16384,"model_type":"chat","tagline":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.74,"output":3.5,"cached_input":0.15},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"moonshotai/kimi-k2","name":"MoonshotAI: Kimi K2 0711","created_by":"moonshotai","source":"official","last_updated":"2026-06-29","description":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.57,"output":2.3},"family":"kimi-k2","model_type":"chat","release_date":"2025-07-11","tagline":"Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass.","tools":["function_calling"],"max_output_tokens":100352},{"id":"morph/morph-v3-fast","name":"Morph: Morph V3 Fast","created_by":"morph","source":"official","last_updated":"2026-03-23","description":"Morph's fastest apply model for code edits.","context_window":81920,"max_output_tokens":38000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.8,"output":1.2},"release_date":"2025-07-07","tagline":"Morph's fastest apply model for code edits."},{"id":"morph/morph-v3-large","name":"Morph: Morph V3 Large","created_by":"morph","source":"official","last_updated":"2026-06-20","description":"Morph's high-accuracy apply model for complex code edits.","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.9,"output":1.9},"release_date":"2025-07-07","tagline":"Morph's high-accuracy apply model for complex code edits.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"nex-agi/deepseek-v3.1-nex-n1","name":"Nex AGI: DeepSeek V3.1 Nex N1","created_by":"nex-agi","source":"official","last_updated":"2026-06-03","family":"deepseek","description":"DeepSeek V3.1 Nex-N1 is the flagship release of the Nex-N1 series — a post-trained model designed to highlight agent autonomy, tool use, and real-world productivity.","context_window":131072,"max_output_tokens":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.135,"output":0.5},"release_date":"2025-12-08","tagline":"DeepSeek V3.1 Nex-N1 is the flagship release of the Nex-N1 series — a post-trained model designed to highlight agent autonomy, tool use, and real-world productivity.","tools":["function_calling"],"model_type":"chat","status":"deprecated","deprecation_date":"2026-06-08"},{"id":"nex-agi/nex-n2-pro:free","name":"Nex AGI: Nex-N2-Pro (free)","created_by":"nex-agi","source":"official","description":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total.","release_date":"2026-06-08","context_window":262144,"max_output_tokens":262144,"tagline":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20","status":"deprecated","deprecation_date":"2026-06-22"},{"id":"nex-agi/nex-n2-pro","name":"Nex AGI: Nex-N2-Pro","created_by":"nex-agi","source":"official","description":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total.","release_date":"2026-06-08","context_window":262144,"max_output_tokens":262144,"tagline":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total.","capabilities":{"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":1,"cached_input":0.025},"last_updated":"2026-06-29"},{"id":"nousresearch/hermes-2-pro-llama-3-8b","name":"NousResearch: Hermes 2 Pro - Llama-3 8B","created_by":"nousresearch","source":"official","last_updated":"2026-05-29","description":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced...","context_window":8192,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.14,"output":0.14},"model_type":"chat","parameters":8,"release_date":"2024-05-27","tagline":"Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mod","status":"deprecated","deprecation_date":"2026-06-05"},{"id":"nousresearch/hermes-3-llama-3.1-405b:free","name":"Nous: Hermes 3 405B Instruct (free)","created_by":"nousresearch","source":"official","last_updated":"2026-04-07","description":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"model_type":"chat","release_date":"2024-08-16","tagline":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren"},{"id":"nousresearch/hermes-3-llama-3.1-405b","name":"Nous: Hermes 3 405B Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-04-07","description":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":1,"output":1},"model_type":"chat","parameters":405,"release_date":"2024-08-16","tagline":"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coheren"},{"id":"nousresearch/hermes-3-llama-3.1-70b","name":"Nous: Hermes 3 70B Instruct","created_by":"nousresearch","source":"official","last_updated":"2026-06-10","description":"Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, rea","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.7,"output":0.7},"model_type":"chat","parameters":70,"release_date":"2024-08-18","tagline":"Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, rea","max_output_tokens":16384},{"id":"nousresearch/hermes-4-405b","name":"Nous: Hermes 4 405B","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","description":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1,"output":3},"model_type":"chat","parameters":405,"release_date":"2025-08-26","tagline":"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research."},{"id":"nousresearch/hermes-4-70b","name":"Nous: Hermes 4 70B","created_by":"nousresearch","source":"official","last_updated":"2026-03-24","description":"Hermes 4 70B is a hybrid reasoning model from Nous Research, built on Meta-Llama-3.1-70B.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.13,"output":0.4},"model_type":"chat","parameters":70,"release_date":"2025-08-26","tagline":"Hermes 4 70B is a hybrid reasoning model from Nous Research, built on Meta-Llama-3.1-70B."},{"id":"nvidia/llama-3.1-nemotron-70b-instruct","name":"NVIDIA: Llama 3.1 Nemotron 70B Instruct","created_by":"nvidia","source":"official","last_updated":"2026-05-07","family":"llama-3.1","description":"NVIDIA's Llama 3.1 Nemotron 70B is a language model designed for generating precise and useful responses.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.2,"output":1.2},"model_type":"chat","parameters":70,"release_date":"2024-10-15","tagline":"NVIDIA's Llama 3.1 Nemotron 70B is a language model designed for generating precise and useful responses.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"nvidia/llama-3.1-nemotron-ultra-253b-v1","name":"NVIDIA: Llama 3.1 Nemotron Ultra 253B v1","created_by":"nvidia","source":"official","last_updated":"2026-03-24","family":"llama-3.1","description":"Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) optimized for advanced reasoning, human-interactive chat, retrieval-augmented generation (RAG), and tool-calling tasks.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":1.8},"model_type":"chat","parameters":253,"release_date":"2025-04-08","tagline":"Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) optimized for advanced reasoning, human-interactive chat, retrieval-augmented generation (RAG), and tool-calling tasks."},{"id":"nvidia/llama-3.3-nemotron-super-49b-v1.5","name":"NVIDIA: Llama 3.3 Nemotron Super 49B V1.5","created_by":"nvidia","source":"official","last_updated":"2026-06-10","family":"llama-3.3","description":"Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.4,"output":0.4},"model_type":"chat","parameters":49,"release_date":"2025-10-10","tagline":"Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context.","tools":["function_calling"],"max_output_tokens":16384},{"id":"nvidia/nemotron-3-nano-30b-a3b:free","name":"NVIDIA: Nemotron 3 Nano 30B A3B (free)","created_by":"nvidia","source":"official","last_updated":"2026-03-23","description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","context_window":256000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-12-14","tagline":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","tools":["function_calling"]},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"NVIDIA: Nemotron 3 Nano 30B A3B","created_by":"nvidia","source":"official","last_updated":"2026-04-25","description":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.05,"output":0.2},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-12-14","tagline":"NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.","tools":["function_calling"],"max_output_tokens":228000},{"id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free","name":"NVIDIA: Nemotron 3 Nano Omni (free)","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems.","release_date":"2026-04-28","context_window":256000,"max_output_tokens":65536,"parameters":30,"active_parameters":3,"model_type":"reasoning","tagline":"NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text","audio","image","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"nvidia/nemotron-3-super-120b-a12b:free","name":"NVIDIA: Nemotron 3 Super (free)","created_by":"nvidia","source":"official","last_updated":"2026-05-17","description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","context_window":1000000,"max_output_tokens":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"model_type":"chat","parameters":120,"active_parameters":12,"release_date":"2026-03-11","tagline":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","tools":["function_calling"]},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"NVIDIA: Nemotron 3 Super","created_by":"nvidia","source":"official","last_updated":"2026-06-29","description":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","context_window":1000000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.085,"output":0.4,"cached_input":0.1},"model_type":"chat","parameters":120,"active_parameters":12,"release_date":"2026-03-11","tagline":"NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications.","tools":["function_calling"],"max_output_tokens":16384},{"id":"nvidia/nemotron-3-ultra-550b-a55b:free","name":"NVIDIA: Nemotron 3 Ultra (free)","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","release_date":"2026-06-04","context_window":1000000,"max_output_tokens":65536,"parameters":550,"active_parameters":55,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"NVIDIA: Nemotron 3 Ultra","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","release_date":"2026-06-04","context_window":1000000,"max_output_tokens":16384,"parameters":550,"active_parameters":55,"model_type":"chat","tagline":"NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":2.2,"cached_input":0.1},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"nvidia/nemotron-3.5-content-safety:free","name":"NVIDIA: Nemotron 3.5 Content Safety (free)","created_by":"nvidia","source":"official","description":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B.","release_date":"2026-06-04","context_window":128000,"max_output_tokens":8192,"model_type":"chat","tagline":"NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B.","capabilities":{"reasoning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-nano-12b-v2-vl:free","name":"NVIDIA: Nemotron Nano 12B 2 VL (free)","created_by":"nvidia","source":"official","last_updated":"2026-03-23","description":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","context_window":128000,"max_output_tokens":128000,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"model_type":"chat","parameters":12,"release_date":"2025-10-28","tagline":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","tools":["function_calling"]},{"id":"nvidia/nemotron-nano-12b-v2-vl","name":"NVIDIA: Nemotron Nano 12B 2 VL","created_by":"nvidia","source":"official","last_updated":"2026-05-01","description":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","context_window":131072,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":12,"release_date":"2025-10-28","tagline":"NVIDIA Nemotron Nano 2 VL is a 12-billion-parameter open multimodal reasoning model designed for video understanding and document intelligence.","max_output_tokens":16384,"status":"deprecated","deprecation_date":"2026-05-07"},{"id":"nvidia/nemotron-nano-9b-v2:free","name":"NVIDIA: Nemotron Nano 9B V2 (free)","created_by":"nvidia","source":"official","last_updated":"2026-03-24","description":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"model_type":"chat","parameters":9,"release_date":"2025-09-05","tagline":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","tools":["function_calling"]},{"id":"nvidia/nemotron-nano-9b-v2","name":"NVIDIA: Nemotron Nano 9B V2","created_by":"nvidia","source":"official","last_updated":"2026-06-10","description":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.04,"output":0.16},"model_type":"chat","parameters":9,"release_date":"2025-09-05","tagline":"NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks.","tools":["function_calling"],"max_output_tokens":16384,"status":"deprecated","deprecation_date":"2026-06-11"},{"id":"openai/gpt-3.5-turbo-0613","name":"OpenAI: GPT-3.5 Turbo (older v0613)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"GPT-3.5 Turbo is OpenAI's fastest model.","context_window":4095,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":true,"streaming":false,"vision":false,"reasoning":false,"batch":true},"pricing":{"input":1,"output":2},"model_type":"chat","release_date":"2024-01-25","tagline":"GPT-3.5 Turbo is OpenAI's fastest model.","tools":["function_calling"],"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-3.5-turbo-16k","name":"OpenAI: GPT-3.5 Turbo 16k","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost.","context_window":16385,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":3,"output":4},"model_type":"chat","release_date":"2023-08-28","tagline":"This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost.","tools":["function_calling"]},{"id":"openai/gpt-3.5-turbo-instruct","name":"OpenAI: GPT-3.5 Turbo Instruct","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations.","context_window":4095,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"fine_tuning":true,"streaming":false,"vision":false,"tool_call":false,"reasoning":false},"pricing":{"input":1.5,"output":2},"model_type":"chat","release_date":"2023-09-28","tagline":"This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations.","knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-3.5-turbo","name":"OpenAI: GPT-3.5 Turbo","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-3.5","description":"GPT-3.5 Turbo is OpenAI's fastest model.","context_window":16385,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":true,"streaming":false,"vision":false,"reasoning":false,"batch":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","release_date":"2023-05-28","tagline":"GPT-3.5 Turbo is OpenAI's fastest model.","tools":["function_calling"],"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-4-0314","name":"OpenAI: GPT-4 (older v0314)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4","description":"GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14.","context_window":8191,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":30,"output":60},"model_type":"chat","release_date":"2023-05-28","tagline":"GPT-4-0314 is the first version of GPT-4 released, with a context length of 8,192 tokens, and was supported until June 14.","tools":["function_calling"]},{"id":"openai/gpt-4-1106-preview","name":"OpenAI: GPT-4 Turbo (older v1106)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4","description":"The latest GPT-4 Turbo model with vision capabilities.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":30},"model_type":"chat","release_date":"2023-11-06","tagline":"The latest GPT-4 Turbo model with vision capabilities.","tools":["function_calling"]},{"id":"openai/gpt-4-turbo-preview","name":"OpenAI: GPT-4 Turbo Preview","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4-turbo","description":"The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":30},"model_type":"chat","release_date":"2024-01-25","tagline":"The preview GPT-4 model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"openai/gpt-4-turbo","name":"OpenAI: GPT-4 Turbo","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4-turbo","description":"The latest GPT-4 Turbo model with vision capabilities.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":10,"output":30},"model_type":"chat","release_date":"2024-04-09","tagline":"The latest GPT-4 Turbo model with vision capabilities.","tools":["function_calling"],"knowledge_cutoff":"2023-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-mini","name":"OpenAI: GPT-4.1 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4.1","description":"GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost.","context_window":1047576,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.4,"output":1.6,"cached_input":0.1},"model_type":"chat","release_date":"2025-04-14","tagline":"GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-nano","name":"OpenAI: GPT-4.1 Nano","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4.1","description":"For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series.","context_window":1047576,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.1,"output":0.4,"cached_input":0.025},"model_type":"chat","release_date":"2025-04-14","tagline":"For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1","name":"OpenAI: GPT-4.1","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4.1","description":"GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning.","context_window":1047576,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":2,"output":8,"cached_input":0.5},"model_type":"chat","release_date":"2025-04-14","tagline":"GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4","name":"OpenAI: GPT-4","created_by":"openai","source":"official","last_updated":"2026-04-07","family":"gpt-4","description":"OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and adva","context_window":8191,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"vision":false,"reasoning":false,"batch":true},"pricing":{"input":30,"output":60},"model_type":"chat","release_date":"2023-05-28","tagline":"OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and adva","tools":["function_calling"],"knowledge_cutoff":"2023-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-2024-05-13","name":"OpenAI: GPT-4o (2024-05-13)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":4096,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":5,"output":15,"cached_input":1.25},"model_type":"chat","release_date":"2024-05-13","alias":"openai/gpt-4o","tagline":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-2024-08-06","name":"OpenAI: GPT-4o (2024-08-06)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2024-08-06","alias":"openai/gpt-4o","tagline":"The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-2024-11-20","name":"OpenAI: GPT-4o (2024-11-20)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2024-11-20","alias":"openai/gpt-4o","tagline":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-audio-preview","name":"OpenAI: GPT-4o Audio","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"The gpt-4o-audio-preview model adds support for audio inputs as prompts.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["audio","text"],"output":["text","audio"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":2.5,"output":10},"model_type":"chat","release_date":"2025-08-15","tagline":"The gpt-4o-audio-preview model adds support for audio inputs as prompts.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o:extended","name":"OpenAI: GPT-4o (extended)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":64000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":6,"output":18},"model_type":"chat","release_date":"2024-05-13","tagline":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","tools":["function_calling"]},{"id":"openai/gpt-4o-mini-2024-07-18","name":"OpenAI: GPT-4o-mini (2024-07-18)","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.075},"model_type":"chat","release_date":"2024-07-18","alias":"openai/gpt-4o-mini","tagline":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini-search-preview","name":"OpenAI: GPT-4o-mini Search Preview","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o mini Search Preview is a specialized model for web search in Chat Completions.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"tool_call":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","release_date":"2025-03-12","tagline":"GPT-4o mini Search Preview is a specialized model for web search in Chat Completions.","knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini","name":"OpenAI: GPT-4o-mini (2024-07-18)","created_by":"openai","source":"official","last_updated":"2026-06-29","family":"gpt-4o","description":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"reasoning":false,"batch":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.075},"model_type":"chat","release_date":"2024-07-18","tagline":"GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs.","tools":["function_calling"],"snapshots":["openai/gpt-4o-mini-2024-07-18"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-search-preview","name":"OpenAI: GPT-4o Search Preview","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-4o","description":"GPT-4o Search Previewis a specialized model for web search in Chat Completions.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"tool_call":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":2.5,"output":10},"model_type":"chat","release_date":"2025-03-12","tagline":"GPT-4o Search Previewis a specialized model for web search in Chat Completions.","knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o","name":"OpenAI: GPT-4o","created_by":"openai","source":"official","last_updated":"2026-06-03","family":"gpt-4o","description":"GPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2024-05-13","tagline":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability.","tools":["function_calling"],"snapshots":["openai/gpt-4o-2024-11-20","openai/gpt-4o-2024-08-06","openai/gpt-4o-2024-05-13"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-chat","name":"OpenAI: GPT-5 Chat","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications."},{"id":"openai/gpt-5-codex","name":"OpenAI: GPT-5 Codex","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5-Codex is a specialized version of GPT-5 optimized for software engineering and coding workflows.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-09-23","tagline":"GPT-5-Codex is a specialized version of GPT-5 optimized for software engineering and coding workflows.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-image-mini","name":"OpenAI: GPT-5 Image Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2.5,"output":2,"cached_input":0.25},"model_type":"chat","release_date":"2025-10-16","tagline":"GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation.","tools":["function_calling"]},{"id":"openai/gpt-5-image","name":"OpenAI: GPT-5 Image","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["image","text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":10,"output":10,"cached_input":1.25},"model_type":"chat","release_date":"2025-10-14","tagline":"[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities.","tools":["function_calling"]},{"id":"openai/gpt-5-mini","name":"OpenAI: GPT-5 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-nano","name":"OpenAI: GPT-5 Nano","created_by":"openai","source":"official","last_updated":"2026-04-12","family":"gpt-5","description":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.05,"output":0.4,"cached_input":0.01},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-pro","name":"OpenAI: GPT-5 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":120},"model_type":"chat","release_date":"2025-10-06","tagline":"GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-chat","name":"OpenAI: GPT-5.1 Chat","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.1","description":"GPT-5.1 Chat (AKA Instant is the fast, lightweight member of the 5.1 family, optimized for low-latency chat while retaining strong general intelligence.","context_window":128000,"max_output_tokens":32000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.25,"output":10,"cached_input":0.13},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1 Chat (AKA Instant is the fast, lightweight member of the 5.1 family, optimized for low-latency chat while retaining strong general intelligence.","tools":["function_calling"]},{"id":"openai/gpt-5.1-codex-max","name":"OpenAI: GPT-5.1-Codex-Max","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.1","description":"GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-12-04","tagline":"GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex-mini","name":"OpenAI: GPT-5.1-Codex-Mini","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.1","description":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","context_window":400000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex","name":"OpenAI: GPT-5.1-Codex","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.1","description":"GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10,"cached_input":0.13},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1","name":"OpenAI: GPT-5.1","created_by":"openai","source":"official","last_updated":"2026-04-12","family":"gpt-5.1","description":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10,"cached_input":0.13},"model_type":"chat","release_date":"2025-11-13","tagline":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5.","tools":["function_calling"],"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2-chat","name":"OpenAI: GPT-5.2 Chat","created_by":"openai","source":"official","last_updated":"2026-05-29","family":"gpt-5.2","description":"GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2025-12-10","tagline":"GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-08-10"},{"id":"openai/gpt-5.2-codex","name":"OpenAI: GPT-5.2-Codex","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.2","description":"GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2026-01-14","tagline":"GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2-pro","name":"OpenAI: GPT-5.2 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.2","description":"GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":21,"output":168},"model_type":"chat","release_date":"2025-12-10","tagline":"GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2","name":"OpenAI: GPT-5.2","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.2","description":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2025-12-10","tagline":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.3-chat","name":"OpenAI: GPT-5.3 Chat","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.3","description":"GPT-5.3 Chat is an update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2026-03-03","tagline":"GPT-5.3 Chat is an update to ChatGPT's most-used model that makes everyday conversations smoother, more useful, and more directly helpful.","tools":["function_calling"]},{"id":"openai/gpt-5.3-codex","name":"OpenAI: GPT-5.3-Codex","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.3","description":"GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","release_date":"2026-02-24","tagline":"GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilitie","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-image-2","name":"OpenAI: GPT-5.4 Image 2","created_by":"openai","source":"official","family":"gpt-5.4","description":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2.","release_date":"2026-04-21","context_window":272000,"max_output_tokens":128000,"model_type":"chat","tagline":"[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text"],"output":["image","text"]},"pricing":{"input":8,"output":15,"cached_input":2},"last_updated":"2026-04-22"},{"id":"openai/gpt-5.4-mini","name":"OpenAI: GPT-5.4 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5,"cached_input":0.075},"model_type":"chat","release_date":"2026-03-17","tagline":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-nano","name":"OpenAI: GPT-5.4 Nano","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25,"cached_input":0.02},"model_type":"chat","release_date":"2026-03-17","tagline":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-pro","name":"OpenAI: GPT-5.4 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks.","context_window":1050000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false},"pricing":{"input":30,"output":180},"model_type":"chat","release_date":"2026-03-05","tagline":"GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4","name":"OpenAI: GPT-5.4","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5.4","description":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system.","context_window":1050000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15,"cached_input":0.25},"model_type":"chat","release_date":"2026-03-05","tagline":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system.","tools":["function_calling"],"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5-pro","name":"OpenAI: GPT-5.5 Pro","created_by":"openai","source":"official","family":"gpt-5.5","description":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads.","release_date":"2026-04-24","context_window":1050000,"max_output_tokens":128000,"model_type":"chat","tagline":"GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":false,"vision":true,"fine_tuning":false,"batch":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":30,"output":180},"tools":["function_calling"],"last_updated":"2026-04-25","knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5","name":"OpenAI: GPT-5.5","created_by":"openai","source":"official","family":"gpt-5.5","description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks.","release_date":"2026-04-24","context_window":1050000,"max_output_tokens":128000,"model_type":"chat","tagline":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"modalities":{"input":["image","text"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-04-25","knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5","name":"OpenAI: GPT-5","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"gpt-5","description":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","context_window":400000,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","release_date":"2025-08-07","tagline":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience.","tools":["function_calling"],"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-audio-mini","name":"OpenAI: GPT Audio Mini","created_by":"openai","source":"official","last_updated":"2026-04-01","family":"gpt-audio","description":"A cost-efficient version of GPT Audio.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","audio"],"output":["text","audio"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true,"streaming":false,"vision":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.6,"output":2.4},"release_date":"2026-01-19","tagline":"A cost-efficient version of GPT Audio.","model_type":"chat","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-audio","name":"OpenAI: GPT Audio","created_by":"openai","source":"official","last_updated":"2026-04-01","family":"gpt-audio","description":"The gpt-audio model is OpenAI's first generally available audio model.","context_window":128000,"max_output_tokens":16384,"modalities":{"input":["text","audio"],"output":["text","audio"]},"capabilities":{"structured_output":true,"json_mode":true,"tool_call":true,"streaming":true,"vision":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":2.5,"output":10},"release_date":"2026-01-19","tagline":"The gpt-audio model is OpenAI's first generally available audio model.","model_type":"chat","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-chat-latest","name":"OpenAI: GPT Chat Latest","created_by":"openai","source":"official","description":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT.","release_date":"2026-05-05","context_window":400000,"max_output_tokens":128000,"tagline":"GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-05-06"},{"id":"openai/gpt-oss-120b:free","name":"OpenAI: gpt-oss-120b (free)","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"model_type":"chat","release_date":"2025-08-05","tagline":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","tools":["function_calling"]},{"id":"openai/gpt-oss-120b","name":"OpenAI: gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-06-29","family":"gpt-oss","description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":false,"vision":false,"batch":true},"pricing":{"input":0.03,"output":0.15},"model_type":"chat","parameters":120,"release_date":"2025-08-05","tagline":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.","tools":["function_calling"],"max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b:free","name":"OpenAI: gpt-oss-20b (free)","created_by":"openai","source":"official","last_updated":"2026-06-18","family":"gpt-oss","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"model_type":"chat","release_date":"2025-08-05","tagline":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","tools":["function_calling"]},{"id":"openai/gpt-oss-20b","name":"OpenAI: gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-06-02","family":"gpt-oss","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":false,"vision":false,"batch":true},"pricing":{"input":0.029,"output":0.14,"cached_input":0.015},"model_type":"chat","parameters":20,"release_date":"2025-08-05","tagline":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-safeguard-20b","name":"OpenAI: gpt-oss-safeguard-20b","created_by":"openai","source":"official","last_updated":"2026-06-20","family":"gpt-oss","description":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b.","context_window":131072,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.075,"output":0.3,"cached_input":0.038},"model_type":"moderation","parameters":20,"release_date":"2025-10-29","tagline":"gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b.","tools":["function_calling"]},{"id":"openai/o1-pro","name":"OpenAI: o1-pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o1","description":"The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"vision":true,"tool_call":true,"streaming":false,"fine_tuning":false,"batch":true},"pricing":{"input":150,"output":600},"model_type":"reasoning","release_date":"2025-03-19","tagline":"The o1 series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o1","name":"OpenAI: o1","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o1","description":"The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":60,"cached_input":7.5},"model_type":"reasoning","release_date":"2024-12-17","tagline":"The latest and strongest model family from OpenAI, o1 is designed to spend more time thinking before responding.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-deep-research","name":"OpenAI: o3 Deep Research","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":10,"output":40,"cached_input":2.5},"model_type":"chat","release_date":"2025-10-10","tagline":"o3-deep-research is OpenAI's advanced model for deep research, designed to tackle complex, multi-step research tasks.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3-mini-high","name":"OpenAI: o3 Mini High","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.55},"model_type":"reasoning","release_date":"2025-02-12","tagline":"OpenAI o3-mini-high is the same model as [o3-mini](/openai/o3-mini) with reasoning_effort set to high.","tools":["function_calling"]},{"id":"openai/o3-mini","name":"OpenAI: o3 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":false,"fine_tuning":false,"batch":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.55},"model_type":"reasoning","release_date":"2025-01-31","tagline":"OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding.","tools":["function_calling"],"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-pro","name":"OpenAI: o3 Pro","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"vision":true,"streaming":false,"fine_tuning":false,"batch":true},"pricing":{"input":20,"output":80},"model_type":"reasoning","release_date":"2025-06-10","tagline":"The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3","name":"OpenAI: o3","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o3","description":"o3 is a well-rounded and powerful model across domains.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":2,"output":8,"cached_input":0.5},"model_type":"reasoning","release_date":"2025-04-16","tagline":"o3 is a well-rounded and powerful model across domains.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o4-mini-deep-research","name":"OpenAI: o4 Mini Deep Research","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o4","description":"o4-mini-deep-research is OpenAI's faster, more affordable deep research model—ideal for tackling complex, multi-step research tasks.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":false,"batch":true},"pricing":{"input":2,"output":8,"cached_input":0.5},"model_type":"chat","release_date":"2025-10-10","tagline":"o4-mini-deep-research is OpenAI's faster, more affordable deep research model—ideal for tackling complex, multi-step research tasks.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o4-mini-high","name":"OpenAI: o4 Mini High","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o4","description":"OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.275},"model_type":"reasoning","release_date":"2025-04-16","tagline":"OpenAI o4-mini-high is the same model as [o4-mini](/openai/o4-mini) with reasoning_effort set to high.","tools":["function_calling"]},{"id":"openai/o4-mini","name":"OpenAI: o4 Mini","created_by":"openai","source":"official","last_updated":"2026-03-24","family":"o4","description":"OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities.","context_window":200000,"max_output_tokens":100000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true,"batch":true},"pricing":{"input":1.1,"output":4.4,"cached_input":0.275},"model_type":"reasoning","release_date":"2025-04-16","tagline":"OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities.","tools":["function_calling"],"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openrouter/auto","name":"Auto Router","created_by":"openrouter","source":"official","last_updated":"2026-05-10","description":"Your prompt will be processed by a meta-model and routed to one of dozens of models (see below), optimizing for the best possible output.","context_window":2000000,"modalities":{"input":["text","image","audio","video"],"output":["text","image"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"release_date":"2023-11-08","tagline":"Your prompt will be processed by a meta-model and routed to one of dozens of models (see below), optimizing for the best possible output.","tools":["function_calling"]},{"id":"openrouter/bodybuilder","name":"Body Builder (beta)","created_by":"openrouter","source":"official","last_updated":"2026-03-23","description":"Transform your natural language requests into structured OpenRouter API request objects.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"release_date":"2025-12-05","tagline":"Transform your natural language requests into structured OpenRouter API request objects."},{"id":"openrouter/elephant-alpha","name":"Elephant","created_by":"openrouter","source":"official","description":"Elephant Alpha is a 100B-parameter text model focused on intelligence efficiency, delivering strong performance while minimizing token usage.","release_date":"2026-04-13","context_window":262144,"max_output_tokens":32768,"tagline":"Elephant Alpha is a 100B-parameter text model focused on intelligence efficiency, delivering strong reasoning performance while minimizing token usage.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-16"},{"id":"openrouter/free","name":"Free Models Router","created_by":"openrouter","source":"official","last_updated":"2026-03-24","description":"The simplest way to get free inference.","context_window":200000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"release_date":"2026-02-01","tagline":"The simplest way to get free inference.","tools":["function_calling"]},{"id":"openrouter/fusion","name":"OpenRouter: Fusion","created_by":"openrouter","source":"official","description":"Fusion turns your prompt into a small multi-model deliberation.","release_date":"2026-06-13","context_window":1000000,"tagline":"Fusion turns your prompt into a small multi-model deliberation.","modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"openrouter/owl-alpha","name":"Owl Alpha","created_by":"openrouter","source":"official","description":"Owl Alpha is a high-performance foundation model designed for agentic workloads.","release_date":"2026-04-28","context_window":1048756,"max_output_tokens":262144,"tagline":"Owl Alpha is a high-performance foundation model designed for agentic workloads.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01"},{"id":"openrouter/pareto-code","name":"Pareto Code Router","created_by":"openrouter","source":"official","description":"The Pareto Router maintains a tiered shortlist of strong coding models, ranked by [Artificial Analysis](https://artificialanalysis.ai/) coding percentiles.","release_date":"2026-04-21","context_window":2000000,"tagline":"The Pareto Router is a way to have OpenRouter always pick a strong coding model for your needs without committing to a specific one.","modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"perceptron/perceptron-mk1","name":"Perceptron: Perceptron Mk1","created_by":"perceptron","source":"official","description":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces","release_date":"2026-05-12","context_window":32768,"max_output_tokens":8192,"tagline":"Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.15,"output":1.5},"last_updated":"2026-05-14"},{"id":"perplexity/sonar-deep-research","name":"Perplexity: Sonar Deep Research","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":2,"output":8},"family":"sonar","model_type":"chat","release_date":"2025-03-07","tagline":"Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics."},{"id":"perplexity/sonar-pro-search","name":"Perplexity: Sonar Pro Search","created_by":"perplexity","source":"official","last_updated":"2026-03-24","description":"Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system.","context_window":200000,"max_output_tokens":8000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15},"family":"sonar","model_type":"chat","release_date":"2025-10-30","tagline":"Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system."},{"id":"perplexity/sonar-pro","name":"Perplexity: Sonar Pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Note: Sonar Pro pricing includes Perplexity search pricing.","context_window":200000,"max_output_tokens":8000,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":3,"output":15},"family":"sonar","model_type":"chat","release_date":"2025-03-07","tagline":"Note: Sonar Pro pricing includes Perplexity search pricing."},{"id":"perplexity/sonar-reasoning-pro","name":"Perplexity: Sonar Reasoning Pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Note: Sonar Pro pricing includes Perplexity search pricing.","context_window":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"reasoning":true},"pricing":{"input":2,"output":8},"model_type":"reasoning","family":"sonar","release_date":"2025-03-07","tagline":"Note: Sonar Pro pricing includes Perplexity search pricing."},{"id":"perplexity/sonar","name":"Perplexity: Sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","description":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources.","context_window":127072,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1,"output":1},"family":"sonar","model_type":"chat","release_date":"2025-01-27","tagline":"Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources."},{"id":"poolside/laguna-m.1:free","name":"Poolside: Laguna M.1 (free)","created_by":"poolside","source":"official","description":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai/), optimized for complex software engineering tasks.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai), optimized for complex software engineering tasks.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"poolside/laguna-m.1","name":"Poolside: Laguna M.1","created_by":"poolside","source":"official","description":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai/), optimized for complex software engineering tasks.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna M.1 is the flagship coding agent model from [Poolside](https://poolside.ai/), optimized for complex software engineering tasks.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.2,"output":0.4,"cached_input":0.1},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"poolside/laguna-xs.2:free","name":"Poolside: Laguna XS.2 (free)","created_by":"poolside","source":"official","description":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai/), their efficient coding agent series.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai), their efficient coding agent series.","capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"poolside/laguna-xs.2","name":"Poolside: Laguna XS.2","created_by":"poolside","source":"official","description":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai/), their efficient coding agent series.","release_date":"2026-04-28","context_window":262144,"max_output_tokens":32768,"tagline":"Laguna XS.2 is the second-generation model in the XS size class from [Poolside](https://poolside.ai/), their efficient coding agent series.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.2,"cached_input":0.05},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"prime-intellect/intellect-3","name":"Prime Intellect: INTELLECT-3","created_by":"prime-intellect","source":"official","last_updated":"2026-03-24","description":"INTELLECT-3 is a 106B-parameter Mixture-of-Experts model (12B active) post-trained from GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":1.1},"release_date":"2025-11-27","tagline":"INTELLECT-3 is a 106B-parameter Mixture-of-Experts model (12B active) post-trained from GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).","tools":["function_calling"]},{"id":"qwen/qwen-2.5-72b-instruct","name":"Qwen2.5 72B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen","description":"Qwen2.5 72B is the latest series of Qwen large language models.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.36,"output":0.4},"model_type":"chat","parameters":72,"release_date":"2024-09-19","tagline":"Qwen2.5 72B is the latest series of Qwen large language models.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen-2.5-7b-instruct","name":"Qwen: Qwen2.5 7B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen","description":"Qwen2.5 7B is the latest series of Qwen large language models.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.04,"output":0.1},"model_type":"chat","parameters":7,"release_date":"2024-10-16","tagline":"Qwen2.5 7B is the latest series of Qwen large language models.","tools":["function_calling"],"max_output_tokens":32768,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen-2.5-coder-32b-instruct","name":"Qwen2.5 Coder 32B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-23","family":"qwen","description":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen).","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.66,"output":1},"model_type":"code","parameters":32,"release_date":"2024-11-11","tagline":"Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen).","max_output_tokens":32768},{"id":"qwen/qwen-2.5-vl-7b-instruct","name":"Qwen: Qwen2.5-VL 7B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen","description":"Qwen2.5 VL 7B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\n- SoTA understanding of images of various resolution & ratio: Qwen2.5-VL achieves state-of-the-art performanc","context_window":32768,"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.2,"output":0.2},"model_type":"chat","parameters":7,"release_date":"2024-08-28","tagline":"Qwen2.5 VL 7B is a multimodal LLM from the Qwen Team with the following key enhancements:\n\n- SoTA understanding of images of various resolution & ratio: Qwen2.5-VL achieves state-of-the-art performanc","license":"apache-2.0","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true}},{"id":"qwen/qwen-max","name":"Qwen: Qwen-Max ","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen-Max, based on Qwen2.5, provides the best inference performance among [Qwen models](/qwen), especially for complex multi-step tasks.","context_window":32768,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true},"pricing":{"input":1.04,"output":4.16,"cached_input":0.208},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen-Max, based on Qwen2.5, provides the best inference performance among [Qwen models](/qwen), especially for complex multi-step tasks.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"qwen/qwen-plus-2025-07-28:thinking","name":"Qwen: Qwen Plus 0728 (thinking)","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","context_window":1000000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.26,"output":0.78},"model_type":"chat","release_date":"2025-09-08","tagline":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","tools":["function_calling"]},{"id":"qwen/qwen-plus-2025-07-28","name":"Qwen: Qwen Plus 0728","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","context_window":1000000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.26,"output":0.78,"cached_input":0.052},"model_type":"chat","release_date":"2025-09-08","alias":"qwen/qwen-plus","tagline":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","tools":["function_calling"]},{"id":"qwen/qwen-plus","name":"Qwen: Qwen-Plus","created_by":"qwen","source":"official","last_updated":"2026-03-26","family":"qwen","description":"Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.","context_window":1000000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true},"pricing":{"input":0.26,"output":0.78,"cached_input":0.052},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen Plus 0728, based on the Qwen3 foundation model, is a 1 million context hybrid reasoning model with a balanced performance, speed, and cost combination.","tools":["function_calling"],"snapshots":["qwen/qwen-plus-2025-07-28"],"license":"proprietary","open_weight":false},{"id":"qwen/qwen-turbo","name":"Qwen: Qwen-Turbo","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen-Turbo, based on Qwen2.5, is a 1M context model that provides fast speed and low cost, suitable for simple tasks.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true},"pricing":{"input":0.033,"output":0.13,"cached_input":0.007},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen-Turbo, based on Qwen2.5, is a 1M context model that provides fast speed and low cost, suitable for simple tasks.","tools":["function_calling"],"license":"proprietary","open_weight":false},{"id":"qwen/qwen-vl-max","name":"Qwen: Qwen VL Max","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen VL Max is a visual understanding model with 7500 tokens context length.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.52,"output":2.08},"model_type":"chat","release_date":"2025-02-01","tagline":"Qwen VL Max is a visual understanding model with 7500 tokens context length.","tools":["function_calling"]},{"id":"qwen/qwen-vl-plus","name":"Qwen: Qwen VL Plus","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen","description":"Qwen's Enhanced Large Visual Language Model.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.137,"output":0.41,"cached_input":0.027},"model_type":"chat","release_date":"2025-02-05","tagline":"Qwen's Enhanced Large Visual Language Model."},{"id":"qwen/qwen2.5-coder-7b-instruct","name":"Qwen: Qwen2.5 Coder 7B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen2.5","description":"Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing.","context_window":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.03,"output":0.09},"model_type":"code","parameters":7,"release_date":"2025-04-15","tagline":"Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing."},{"id":"qwen/qwen2.5-vl-32b-instruct","name":"Qwen: Qwen2.5 VL 32B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen2.5","description":"Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities.","context_window":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"tool_call":true,"batch":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":32,"release_date":"2025-03-24","tagline":"Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities.","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen2.5-vl-72b-instruct","name":"Qwen: Qwen2.5 VL 72B Instruct","created_by":"qwen","source":"official","last_updated":"2026-06-18","family":"qwen2.5","description":"Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects.","context_window":131072,"max_output_tokens":128000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"tool_call":true,"batch":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.8,"output":1,"cached_input":0.4},"model_type":"chat","parameters":72,"release_date":"2025-02-01","tagline":"Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects.","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-14b","name":"Qwen: Qwen3 14B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue.","context_window":131702,"max_output_tokens":40960,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.24},"model_type":"chat","parameters":14,"release_date":"2025-04-28","tagline":"Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-235b-a22b-2507","name":"Qwen: Qwen3 235B A22B Instruct 2507","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.09,"output":0.1},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-07-21","max_output_tokens":16384,"alias":"qwen/qwen3-235b-a22b","tagline":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","tools":["function_calling"]},{"id":"qwen/qwen3-235b-a22b-thinking-2507","name":"Qwen: Qwen3 235B A22B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-06-02","family":"qwen3","description":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.1,"cached_input":0.1},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-07-25","alias":"qwen/qwen3-235b-a22b-thinking","tagline":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","tools":["function_calling"],"max_output_tokens":262144,"status":"deprecated","deprecation_date":"2026-04-22","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-235b-a22b-thinking","name":"Qwen: Qwen3 235B A22B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-06-02","family":"qwen3","description":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","release_date":"2025-07-25","context_window":262144,"parameters":235,"active_parameters":22,"model_type":"chat","tagline":"Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.1,"cached_input":0.1},"tools":["function_calling"],"snapshots":["qwen/qwen3-235b-a22b-thinking-2507"],"max_output_tokens":262144,"status":"deprecated","deprecation_date":"2026-04-22","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-235b-a22b","name":"Qwen: Qwen3 235B A22B","created_by":"qwen","source":"official","last_updated":"2026-03-26","family":"qwen3","description":"Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.455,"output":1.82},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-04-28","tagline":"Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass.","tools":["function_calling"],"snapshots":["qwen/qwen3-235b-a22b-2507"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-instruct-2507","name":"Qwen: Qwen3 30B A3B Instruct 2507","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","context_window":131072,"max_output_tokens":32000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.048,"output":0.193},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-07-29","alias":"qwen/qwen3-30b-a3b-instruct","tagline":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-instruct","name":"Qwen: Qwen3 30B A3B Instruct 2507","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","release_date":"2025-07-29","context_window":131072,"max_output_tokens":32000,"parameters":30,"active_parameters":3,"model_type":"chat","tagline":"Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.048,"output":0.193},"tools":["function_calling"],"snapshots":["qwen/qwen3-30b-a3b-instruct-2507"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-thinking-2507","name":"Qwen: Qwen3 30B A3B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.08,"output":0.4,"cached_input":0.08},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-08-28","alias":"qwen/qwen3-30b-a3b-thinking","tagline":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b-thinking","name":"Qwen: Qwen3 30B A3B Thinking 2507","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","release_date":"2025-08-28","context_window":131072,"max_output_tokens":131072,"parameters":30,"active_parameters":3,"model_type":"chat","tagline":"Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.08,"output":0.4,"cached_input":0.08},"tools":["function_calling"],"snapshots":["qwen/qwen3-30b-a3b-thinking-2507"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-30b-a3b","name":"Qwen: Qwen3 30B A3B","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","description":"Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.12,"output":0.5},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-04-28","tagline":"Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tas","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-06-05","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-32b","name":"Qwen: Qwen3 32B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.08,"output":0.28,"cached_input":0.04},"model_type":"chat","parameters":32,"release_date":"2025-04-28","tagline":"Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-4b:free","name":"Qwen: Qwen3 4B (free)","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-4B is a 4 billion parameter dense language model from the Qwen3 series, designed to support both general-purpose and reasoning-intensive tasks.","status":"deprecated","deprecation_date":"2026-03-29","context_window":40960,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"model_type":"chat","release_date":"2025-04-30","tagline":"Qwen3-4B is a 4 billion parameter dense language model from the Qwen3 series, designed to support both general-purpose and reasoning-intensive tasks.","tools":["function_calling"]},{"id":"qwen/qwen3-8b","name":"Qwen: Qwen3 8B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue.","context_window":131072,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.05,"output":0.4,"cached_input":0.05},"model_type":"chat","parameters":8,"release_date":"2025-04-28","tagline":"Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder-30b-a3b-instruct","name":"Qwen: Qwen3 Coder 30B A3B Instruct","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.","context_window":160000,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.07,"output":0.27},"model_type":"code","parameters":30,"active_parameters":3,"release_date":"2025-07-31","tagline":"Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, an","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder-flash","name":"Qwen: Qwen3 Coder Flash","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.195,"output":0.975,"cached_input":0.039},"model_type":"code","release_date":"2025-09-17","tagline":"Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder:free","name":"Qwen: Qwen3 Coder 480B A35B (free)","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","context_window":1048576,"max_output_tokens":262000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"model_type":"code","parameters":480,"active_parameters":35,"release_date":"2025-07-23","tagline":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","tools":["function_calling"]},{"id":"qwen/qwen3-coder-next","name":"Qwen: Qwen3 Coder Next","created_by":"qwen","source":"official","last_updated":"2026-05-07","family":"qwen3","description":"Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.11,"output":0.8,"cached_input":0.07},"model_type":"code","release_date":"2026-02-04","tagline":"Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder-plus","name":"Qwen: Qwen3 Coder Plus","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true,"reasoning":true,"batch":true},"pricing":{"input":0.65,"output":3.25,"cached_input":0.13},"model_type":"code","release_date":"2025-09-23","tagline":"Qwen3 Coder Plus is Alibaba's proprietary version of the Open Source Qwen3 Coder 480B A35B.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-coder","name":"Qwen: Qwen3 Coder 480B A35B","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","context_window":1048576,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.22,"output":1.8,"cached_input":0.022},"model_type":"code","parameters":480,"active_parameters":35,"release_date":"2025-07-23","tagline":"Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.","tools":["function_calling"],"max_output_tokens":65536},{"id":"qwen/qwen3-max-thinking","name":"Qwen: Qwen3 Max Thinking","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.78,"output":3.9},"model_type":"chat","release_date":"2026-02-09","tagline":"Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning.","tools":["function_calling"]},{"id":"qwen/qwen3-max","name":"Qwen: Qwen3 Max","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.78,"output":3.9,"cached_input":0.156},"model_type":"chat","release_date":"2025-09-23","tagline":"Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the Janua","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-instruct:free","name":"Qwen: Qwen3 Next 80B A3B Instruct (free)","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"model_type":"chat","parameters":80,"active_parameters":3,"release_date":"2025-09-11","tagline":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","tools":["function_calling"]},{"id":"qwen/qwen3-next-80b-a3b-instruct","name":"Qwen: Qwen3 Next 80B A3B Instruct","created_by":"qwen","source":"official","last_updated":"2026-04-25","family":"qwen3","description":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","context_window":262144,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"reasoning":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.09,"output":1.1},"model_type":"chat","parameters":80,"active_parameters":3,"release_date":"2025-09-11","tagline":"Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces.","tools":["function_calling"],"max_output_tokens":16384,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-next-80b-a3b-thinking","name":"Qwen: Qwen3 Next 80B A3B Thinking","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"pricing":{"input":0.098,"output":0.78},"model_type":"chat","parameters":80,"active_parameters":3,"release_date":"2025-09-11","tagline":"Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-235b-a22b-instruct","name":"Qwen: Qwen3 VL 235B A22B Instruct","created_by":"qwen","source":"official","last_updated":"2026-04-25","family":"qwen3","description":"Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video.","context_window":262144,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.88,"cached_input":0.11},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-09-23","tagline":"Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video.","tools":["function_calling"],"max_output_tokens":16384,"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-235b-a22b-thinking","name":"Qwen: Qwen3 VL 235B A22B Thinking","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.26,"output":2.6},"model_type":"chat","parameters":235,"active_parameters":22,"release_date":"2025-09-23","tagline":"Qwen3-VL-235B-A22B Thinking is a multimodal model that unifies strong text generation with visual understanding across images and video.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-30b-a3b-instruct","name":"Qwen: Qwen3 VL 30B A3B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.13,"output":0.52},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-10-06","tagline":"Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-30b-a3b-thinking","name":"Qwen: Qwen3 VL 30B A3B Thinking","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3","description":"Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.13,"output":1.56},"model_type":"chat","parameters":30,"active_parameters":3,"release_date":"2025-10-06","tagline":"Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-32b-instruct","name":"Qwen: Qwen3 VL 32B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video.","context_window":262144,"max_output_tokens":32768,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.104,"output":0.416},"model_type":"chat","parameters":32,"release_date":"2025-10-23","tagline":"Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-8b-instruct","name":"Qwen: Qwen3 VL 8B Instruct","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video.","context_window":256000,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.08,"output":0.5},"model_type":"chat","parameters":8,"release_date":"2025-10-14","tagline":"Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3-vl-8b-thinking","name":"Qwen: Qwen3 VL 8B Thinking","created_by":"qwen","source":"official","last_updated":"2026-05-17","family":"qwen3","description":"Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences.","context_window":256000,"max_output_tokens":32768,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.117,"output":1.365},"model_type":"chat","parameters":8,"release_date":"2025-10-14","tagline":"Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-122b-a10b","name":"Qwen: Qwen3.5-122B-A10B","created_by":"qwen","source":"official","last_updated":"2026-05-23","family":"qwen3.5","description":"The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":262144,"max_output_tokens":262144,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.26,"output":2.08},"model_type":"chat","parameters":122,"active_parameters":10,"release_date":"2026-02-25","tagline":"The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference eff","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-27b","name":"Qwen: Qwen3.5-27B","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","description":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","context_window":262144,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.195,"output":1.56},"model_type":"chat","parameters":27,"release_date":"2026-02-25","tagline":"The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance.","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-35b-a3b","name":"Qwen: Qwen3.5-35B-A3B","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3.5","description":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":262144,"max_output_tokens":81920,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.14,"output":1,"cached_input":0.05},"model_type":"chat","parameters":35,"active_parameters":3,"release_date":"2026-02-25","tagline":"The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inf","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-397b-a17b","name":"Qwen: Qwen3.5 397B A17B","created_by":"qwen","source":"official","last_updated":"2026-06-18","family":"qwen3.5","description":"The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":256000,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"pricing":{"input":0.385,"output":2.45,"cached_input":0.195},"model_type":"chat","parameters":397,"active_parameters":17,"release_date":"2026-02-16","tagline":"The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher infere","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.5-9b","name":"Qwen: Qwen3.5-9B","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3.5","description":"Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture.","context_window":262144,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.1,"output":0.15},"model_type":"chat","parameters":9,"release_date":"2026-03-10","tagline":"Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture.","tools":["function_calling"],"max_output_tokens":262144},{"id":"qwen/qwen3.5-flash-02-23","name":"Qwen: Qwen3.5-Flash","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","description":"The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.065,"output":0.26},"model_type":"chat","release_date":"2026-02-25","tagline":"The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference effic","tools":["function_calling"]},{"id":"qwen/qwen3.5-plus-02-15","name":"Qwen: Qwen3.5 Plus 2026-02-15","created_by":"qwen","source":"official","last_updated":"2026-03-24","family":"qwen3.5","description":"The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference efficiency.","context_window":1000000,"max_output_tokens":65536,"modalities":{"input":["text","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.26,"output":1.56},"model_type":"chat","release_date":"2026-02-16","tagline":"The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference e","tools":["function_calling"]},{"id":"qwen/qwen3.5-plus-20260420","name":"Qwen: Qwen3.5 Plus 2026-04-20","created_by":"qwen","source":"official","family":"qwen3.5","description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","alias":"qwen/qwen3.5-plus","tagline":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.3,"output":1.8},"tools":["function_calling"],"last_updated":"2026-05-17"},{"id":"qwen/qwen3.5-plus","name":"Qwen: Qwen3.5 Plus 2026-04-20","created_by":"qwen","source":"official","family":"qwen3.5","description":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.3,"output":1.8},"tools":["function_calling"],"snapshots":["qwen/qwen3.5-plus-20260420"],"last_updated":"2026-05-17","license":"proprietary","open_weight":false},{"id":"qwen/qwen3.6-27b","name":"Qwen: Qwen3.6 27B","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":262140,"parameters":27,"model_type":"chat","tagline":"Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.26,"output":2.385},"tools":["function_calling"],"last_updated":"2026-06-29","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-35b-a3b","name":"Qwen: Qwen3.6 35B A3B","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":262144,"parameters":35,"active_parameters":3,"model_type":"chat","tagline":"Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true,"tool_call":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.14,"output":1,"cached_input":0.05},"last_updated":"2026-06-20","tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-flash","name":"Qwen: Qwen3.6 Flash","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series.","release_date":"2026-04-27","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.188,"output":1.125},"tools":["function_calling"],"last_updated":"2026-05-17","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-max-preview","name":"Qwen: Qwen3.6 Max Preview","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters.","release_date":"2026-04-27","context_window":262144,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.04,"output":6.24},"tools":["function_calling"],"last_updated":"2026-05-01","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.6-plus:free","name":"Qwen: Qwen3.6 Plus (free)","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","release_date":"2026-04-02","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"qwen/qwen3.6-plus-preview:free","name":"Qwen: Qwen3.6 Plus Preview (free)","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen 3.6 Plus Preview is the next-generation evolution of the Qwen Plus series, featuring an advanced hybrid architecture that improves efficiency and scalability.","release_date":"2026-03-30","context_window":1000000,"max_output_tokens":32000,"model_type":"chat","tagline":"Qwen 3.6 Plus Preview is the next-generation evolution of the Qwen Plus series, featuring an advanced hybrid architecture that improves efficiency and scalability.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"qwen/qwen3.6-plus","name":"Qwen: Qwen3.6 Plus","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","release_date":"2026-04-02","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"batch":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.325,"output":1.95},"tools":["function_calling"],"last_updated":"2026-04-08","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.7-max","name":"Qwen: Qwen3.7 Max","created_by":"qwen","source":"official","family":"qwen3.7","description":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series.","release_date":"2026-05-21","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.25,"output":3.75,"cached_input":0.25},"tools":["function_calling"],"last_updated":"2026-05-29","license":"apache-2.0","open_weight":true},{"id":"qwen/qwen3.7-plus","name":"Qwen: Qwen3.7 Plus","created_by":"qwen","source":"official","family":"qwen3.7","description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series.","release_date":"2026-06-03","context_window":1000000,"max_output_tokens":65536,"model_type":"chat","tagline":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.32,"output":1.28,"cached_input":0.064},"tools":["function_calling"],"last_updated":"2026-06-18","license":"apache-2.0","open_weight":true},{"id":"qwen/qwq-32b","name":"Qwen: QwQ 32B","created_by":"qwen","source":"official","last_updated":"2026-04-22","family":"qwq","description":"QwQ is the reasoning model of the Qwen series.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.58},"model_type":"chat","parameters":32,"release_date":"2025-03-05","tagline":"QwQ is the reasoning model of the Qwen series.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-04-29","license":"apache-2.0","open_weight":true},{"id":"reka/reka-edge","name":"Reka Edge","created_by":"reka","source":"official","description":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","release_date":"2026-03-20","context_window":16384,"max_output_tokens":16384,"tagline":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"vision":true,"streaming":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"tools":["function_calling"],"last_updated":"2026-03-27","license":"proprietary","open_weight":false},{"id":"rekaai/reka-edge","name":"Reka Edge","created_by":"rekaai","source":"official","description":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","release_date":"2026-03-20","context_window":16384,"max_output_tokens":16384,"tagline":"Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs.","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":0.1,"output":0.1},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"rekaai/reka-flash-3","name":"Reka Flash 3","created_by":"rekaai","source":"official","description":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka.","release_date":"2025-03-12","context_window":65536,"max_output_tokens":65536,"tagline":"Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka.","capabilities":{"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.2},"last_updated":"2026-06-20"},{"id":"relace/relace-apply-3","name":"Relace: Relace Apply 3","created_by":"relace","source":"official","last_updated":"2026-03-23","description":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files.","context_window":256000,"max_output_tokens":128000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.85,"output":1.25},"release_date":"2025-09-26","tagline":"Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files."},{"id":"relace/relace-search","name":"Relace: Relace Search","created_by":"relace","source":"official","last_updated":"2026-06-20","description":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request.","context_window":256000,"max_output_tokens":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":1,"output":3},"model_type":"chat","release_date":"2025-12-08","tagline":"The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request.","tools":["function_calling"]},{"id":"sakana/fugu-ultra","name":"Sakana: Fugu Ultra","created_by":"sakana","source":"official","description":"Fugu Ultra is the higher-performance model in Sakana AI's Fugu family.","release_date":"2026-06-24","context_window":1000000,"max_output_tokens":128000,"tagline":"Fugu Ultra is the higher-performance model in Sakana AI's Fugu family.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5,"output":30,"cached_input":0.5},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"sao10k/l3-euryale-70b","name":"Sao10k: Llama 3 Euryale 70B v2.1","created_by":"sao10k","source":"official","last_updated":"2026-05-29","description":"Euryale 70B v2.1 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","context_window":8192,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":1.48,"output":1.48},"parameters":70,"release_date":"2024-06-18","tagline":"Euryale 70B v2.1 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-06-05"},{"id":"sao10k/l3-lunaris-8b","name":"Sao10K: Llama 3 8B Lunaris","created_by":"sao10k","source":"official","last_updated":"2026-04-25","description":"Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3.","context_window":8192,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.04,"output":0.05},"parameters":8,"release_date":"2024-08-13","tagline":"Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3.","max_output_tokens":16384},{"id":"sao10k/l3.1-70b-hanami-x1","name":"Sao10K: Llama 3.1 70B Hanami x1","created_by":"sao10k","source":"official","last_updated":"2026-06-20","description":"This is [Sao10K](/sao10k)'s experiment over [Euryale v2.2](/sao10k/l3.1-euryale-70b).","context_window":16000,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":3,"output":3},"parameters":70,"release_date":"2025-01-08","tagline":"This is [Sao10K](/sao10k)'s experiment over [Euryale v2.2](/sao10k/l3.1-euryale-70b).","capabilities":{"structured_output":true,"json_mode":true}},{"id":"sao10k/l3.1-euryale-70b","name":"Sao10K: Llama 3.1 Euryale 70B v2.2","created_by":"sao10k","source":"official","last_updated":"2026-03-24","description":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.85,"output":0.85},"parameters":70,"release_date":"2024-08-28","tagline":"Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","tools":["function_calling"]},{"id":"sao10k/l3.3-euryale-70b","name":"Sao10K: Llama 3.3 Euryale 70B","created_by":"sao10k","source":"official","last_updated":"2026-03-24","description":"Euryale L3.3 70B is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k).","context_window":131072,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.65,"output":0.75},"parameters":70,"release_date":"2024-12-18","tagline":"Euryale L3.3 70B is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k)."},{"id":"stepfun/step-3.5-flash:free","name":"StepFun: Step 3.5 Flash (free)","created_by":"stepfun","source":"official","last_updated":"2026-03-23","description":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","context_window":256000,"max_output_tokens":256000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"release_date":"2026-01-29","tagline":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","tools":["function_calling"]},{"id":"stepfun/step-3.5-flash","name":"StepFun: Step 3.5 Flash","created_by":"stepfun","source":"official","last_updated":"2026-06-29","description":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","context_window":262144,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.09,"output":0.3,"cached_input":0.02},"release_date":"2026-01-29","tagline":"Step 3.5 Flash is StepFun's most capable open-source foundation model.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-06-29","license":"apache-2.0","open_weight":true},{"id":"stepfun/step-3.7-flash","name":"StepFun: Step 3.7 Flash","created_by":"stepfun","source":"official","description":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model.","release_date":"2026-05-28","context_window":256000,"max_output_tokens":256000,"tagline":"Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.2,"output":1.15,"cached_input":0.04},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"switchpoint/router","name":"Switchpoint Router","created_by":"switchpoint","source":"official","last_updated":"2026-06-20","description":"Switchpoint AI's router instantly analyzes your request and directs it to the optimal AI from an ever-evolving library.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.85,"output":3.4},"release_date":"2025-07-11","tagline":"Switchpoint AI's router instantly analyzes your request and directs it to the optimal AI from an ever-evolving library."},{"id":"tencent/hunyuan-a13b-instruct","name":"Tencent: Hunyuan A13B Instruct","created_by":"tencent","source":"official","last_updated":"2026-03-24","description":"Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.14,"output":0.57},"release_date":"2025-07-08","tagline":"Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought."},{"id":"tencent/hy3-preview:free","name":"Tencent: Hy3 preview (free)","created_by":"tencent","source":"official","description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","release_date":"2026-04-22","context_window":262144,"max_output_tokens":262144,"tagline":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","status":"deprecated","deprecation_date":"2026-05-08"},{"id":"tencent/hy3-preview","name":"Tencent: Hy3 preview","created_by":"tencent","source":"official","description":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","release_date":"2026-04-22","context_window":262144,"max_output_tokens":262144,"tagline":"Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use.","capabilities":{"tool_call":true,"reasoning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.063,"output":0.21,"cached_input":0.021},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"thedrummer/cydonia-24b-v4.1","name":"TheDrummer: Cydonia 24B V4.1","created_by":"thedrummer","source":"official","last_updated":"2026-06-18","description":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","context_window":131072,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.3,"output":0.5,"cached_input":0.15},"parameters":24,"release_date":"2025-09-27","tagline":"Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"thedrummer/rocinante-12b","name":"TheDrummer: Rocinante 12B","created_by":"thedrummer","source":"official","last_updated":"2026-06-29","description":"Rocinante 12B is designed for engaging storytelling and rich prose.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.25,"output":0.5},"parameters":12,"release_date":"2024-09-30","tagline":"Rocinante 12B is designed for engaging storytelling and rich prose.","tools":["function_calling"]},{"id":"thedrummer/skyfall-36b-v2","name":"TheDrummer: Skyfall 36B V2","created_by":"thedrummer","source":"official","last_updated":"2026-06-20","description":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.55,"output":0.8,"cached_input":0.25},"parameters":36,"release_date":"2025-03-10","tagline":"Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.","capabilities":{"structured_output":true,"json_mode":true}},{"id":"thedrummer/unslopnemo-12b","name":"TheDrummer: UnslopNemo 12B","created_by":"thedrummer","source":"official","last_updated":"2026-03-24","description":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","context_window":32768,"max_output_tokens":32768,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.4,"output":0.4},"parameters":12,"release_date":"2024-11-08","tagline":"UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.","tools":["function_calling"]},{"id":"tngtech/deepseek-r1t2-chimera","name":"TNG: DeepSeek R1T2 Chimera","created_by":"tngtech","source":"official","last_updated":"2026-03-24","family":"deepseek-r1","description":"DeepSeek-TNG-R1T2-Chimera is the second-generation Chimera model from TNG Tech.","context_window":163840,"max_output_tokens":163840,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.3,"output":1.1,"cached_input":0.15},"model_type":"reasoning","release_date":"2025-07-08","tagline":"DeepSeek-TNG-R1T2-Chimera is the second-generation Chimera model from TNG Tech.","tools":["function_calling"]},{"id":"undi95/remm-slerp-l2-13b","name":"ReMM SLERP 13B","created_by":"undi95","source":"official","last_updated":"2026-03-24","description":"A recreation trial of the original MythoMax-L2-B13 but with updated models.","context_window":6144,"max_output_tokens":4096,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"structured_output":true,"json_mode":true},"pricing":{"input":0.45,"output":0.65},"parameters":13,"release_date":"2023-07-22","tagline":"A recreation trial of the original MythoMax-L2-B13 but with updated models."},{"id":"upstage/solar-pro-3","name":"Upstage: Solar Pro 3","created_by":"upstage","source":"official","last_updated":"2026-03-24","description":"Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.15,"output":0.6,"cached_input":0.015},"release_date":"2026-01-27","tagline":"Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model.","tools":["function_calling"]},{"id":"writer/palmyra-x5","name":"Writer: Palmyra X5","created_by":"writer","source":"official","last_updated":"2026-03-23","description":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise.","context_window":1040000,"max_output_tokens":8192,"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":6},"model_type":"chat","release_date":"2026-01-21","tagline":"Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise.","license":"proprietary","open_weight":false,"capabilities":{"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true}},{"id":"x-ai/grok-3-beta","name":"xAI: Grok 3 Beta","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"Grok 3 is the latest model from xAI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.75},"model_type":"chat","release_date":"2025-04-09","tagline":"Grok 3 is the latest model from xAI.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-3-mini-beta","name":"xAI: Grok 3 Mini Beta","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"Grok 3 Mini is a lightweight, smaller thinking model.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.3,"output":0.5,"cached_input":0.075},"model_type":"chat","release_date":"2025-04-09","tagline":"Grok 3 Mini is a lightweight, smaller thinking model.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-3-mini","name":"xAI: Grok 3 Mini","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"A lightweight model that thinks before responding.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true},"pricing":{"input":0.3,"output":0.5,"cached_input":0.075},"model_type":"chat","release_date":"2025-06-10","tagline":"A lightweight model that thinks before responding.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"x-ai/grok-3","name":"xAI: Grok 3","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-3","description":"Grok 3 is the latest model from xAI.","context_window":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.75},"model_type":"chat","release_date":"2025-06-10","tagline":"Grok 3 is the latest model from xAI.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"x-ai/grok-4-fast","name":"xAI: Grok 4 Fast","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-4","description":"Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window.","context_window":2000000,"max_output_tokens":30000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5,"cached_input":0.05},"model_type":"chat","release_date":"2025-09-19","tagline":"Grok 4 Fast is xAI's latest multimodal model with SOTA cost-efficiency and a 2M token context window.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-4.1-fast","name":"xAI: Grok 4.1 Fast","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-4.1","description":"Grok 4.1 Fast is xAI's best agentic tool calling model that shines in real-world use cases like customer support and deep research.","context_window":2000000,"max_output_tokens":30000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5,"cached_input":0.05},"model_type":"chat","release_date":"2025-11-19","tagline":"Grok 4.1 Fast is xAI's best agentic tool calling model that shines in real-world use cases like customer support and deep research.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-4.20-beta","name":"xAI: Grok 4.20 Beta","created_by":"xai","source":"official","last_updated":"2026-03-24","family":"grok-4.20","description":"Grok 4.20 Beta is xAI's newest flagship model with industry-leading speed and agentic tool calling capabilities.","context_window":2000000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2026-03-12","tagline":"Grok 4.20 Beta is xAI's newest flagship model with industry-leading speed and agentic tool calling capabilities.","tools":["function_calling"]},{"id":"x-ai/grok-4.20-multi-agent-beta","name":"xAI: Grok 4.20 Multi-Agent Beta","created_by":"xai","source":"official","last_updated":"2026-03-24","family":"grok-4.20","description":"Grok 4.20 Multi-Agent Beta is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows.","context_window":2000000,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":2,"output":6,"cached_input":0.2},"model_type":"chat","release_date":"2026-03-12","tagline":"Grok 4.20 Multi-Agent Beta is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows."},{"id":"x-ai/grok-4.20-multi-agent","name":"xAI: Grok 4.20 Multi-Agent","created_by":"xai","source":"official","family":"grok-4.20","description":"Grok 4.20 Multi-Agent is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows.","release_date":"2026-03-31","context_window":2000000,"model_type":"chat","tagline":"Grok 4.20 Multi-Agent is a variant of xAI’s Grok 4.20 designed for collaborative, agent-based workflows.","capabilities":{"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"last_updated":"2026-06-18"},{"id":"x-ai/grok-4.20","name":"xAI: Grok 4.20","created_by":"xai","source":"official","family":"grok-4.20","description":"Grok 4.20 is a reasoning model from xAI with industry-leading speed and agentic tool calling capabilities.","release_date":"2026-03-31","context_window":2000000,"model_type":"chat","tagline":"Grok 4.20 is xAI's newest flagship model with industry-leading speed and agentic tool calling capabilities.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-10"},{"id":"x-ai/grok-4.3","name":"xAI: Grok 4.3","created_by":"xai","source":"official","family":"grok-4.3","description":"Grok 4.3 is a reasoning model from xAI.","release_date":"2026-04-30","context_window":1000000,"model_type":"chat","tagline":"Grok 4.3 is a reasoning model from xAI.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-01","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"x-ai/grok-4","name":"xAI: Grok 4","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-4","description":"Grok 4 is xAI's latest reasoning model with a 256k context window.","context_window":256000,"modalities":{"input":["image","text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":3,"output":15,"cached_input":0.75},"model_type":"chat","release_date":"2025-07-09","tagline":"Grok 4 is xAI's latest reasoning model with a 256k context window.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15"},{"id":"x-ai/grok-build-0.1","name":"xAI: Grok Build 0.1","created_by":"xai","source":"official","description":"Grok Build 0.1 is xAI’s fast coding model trained specifically for agentic software engineering workflows.","release_date":"2026-05-20","context_window":256000,"model_type":"chat","tagline":"Grok Build 0.1 is xAI’s fast coding model trained specifically for agentic software engineering workflows.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1,"output":2,"cached_input":0.2},"tools":["function_calling"],"last_updated":"2026-05-23","license":"proprietary","open_weight":false},{"id":"x-ai/grok-code-fast-1","name":"xAI: Grok Code Fast 1","created_by":"xai","source":"official","last_updated":"2026-05-07","family":"grok-code","description":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","context_window":256000,"max_output_tokens":10000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"batch":true},"pricing":{"input":0.2,"output":1.5,"cached_input":0.02},"model_type":"code","release_date":"2025-08-26","tagline":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-15","license":"proprietary","open_weight":false},{"id":"xiaomi/mimo-v2-flash","name":"Xiaomi: MiMo-V2-Flash","created_by":"xiaomi","source":"official","last_updated":"2026-05-10","description":"MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi.","context_window":262144,"max_output_tokens":65536,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3,"cached_input":0.01},"release_date":"2025-12-14","tagline":"MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi.","tools":["function_calling"],"family":"mimo","model_type":"chat","knowledge_cutoff":"2024-12","license":"mit","parameters":309,"active_parameters":15,"architecture":"moe","open_weight":true},{"id":"xiaomi/mimo-v2-omni","name":"Xiaomi: MiMo-V2-Omni","created_by":"xiaomi","source":"official","last_updated":"2026-05-29","description":"MiMo-V2-Omni is a frontier omni-modal model that natively processes image, video, and audio inputs within a unified architecture.","context_window":262144,"max_output_tokens":65536,"modalities":{"input":["text","audio","image","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"pricing":{"input":0.4,"output":2,"cached_input":0.08},"release_date":"2026-03-18","tagline":"MiMo-V2-Omni is a frontier omni-modal model that natively processes image, video, and audio inputs within a unified architecture.","tools":["function_calling"],"family":"mimo","model_type":"chat","status":"deprecated","deprecation_date":"2026-05-31","license":"proprietary","open_weight":false},{"id":"xiaomi/mimo-v2-pro","name":"Xiaomi: MiMo-V2-Pro","created_by":"xiaomi","source":"official","last_updated":"2026-05-29","description":"MiMo-V2-Pro is Xiaomi's flagship foundation model, featuring over 1T total parameters and a 1M context length, deeply optimized for agentic scenarios.","context_window":1048576,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"pricing":{"input":1,"output":3,"cached_input":0.2},"release_date":"2026-03-18","tagline":"MiMo-V2-Pro is Xiaomi's flagship foundation model, featuring over 1T total parameters and a 1M context length, deeply optimized for agentic scenarios.","tools":["function_calling"],"family":"mimo","model_type":"chat","status":"deprecated","deprecation_date":"2026-05-31","license":"proprietary","parameters":1000,"active_parameters":42,"architecture":"moe","open_weight":false},{"id":"xiaomi/mimo-v2.5-pro","name":"Xiaomi: MiMo-V2.5-Pro","created_by":"xiaomi","source":"official","family":"mimo","description":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as C","release_date":"2026-04-22","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as C","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.435,"output":0.87,"cached_input":0.004},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"xiaomi/mimo-v2.5","name":"Xiaomi: MiMo-V2.5","created_by":"xiaomi","source":"official","family":"mimo","description":"MiMo-V2.5 is a native omnimodal model by Xiaomi.","release_date":"2026-04-22","context_window":1048576,"max_output_tokens":131072,"model_type":"chat","tagline":"MiMo-V2.5 is a native omnimodal model by Xiaomi.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"vision":true},"modalities":{"input":["text","audio","image","video"],"output":["text"]},"pricing":{"input":0.105,"output":0.28,"cached_input":0.003},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"z-ai/glm-4-32b","name":"Z.ai: GLM 4 32B ","created_by":"z-ai","source":"official","last_updated":"2026-03-23","description":"GLM 4 32B is a cost-effective foundation language model.","context_window":128000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true},"pricing":{"input":0.1,"output":0.1},"family":"glm-4","model_type":"chat","parameters":32,"release_date":"2025-07-24","tagline":"GLM 4 32B is a cost-effective foundation language model.","tools":["function_calling"]},{"id":"z-ai/glm-4.5-air:free","name":"Z.ai: GLM 4.5 Air (free)","created_by":"z-ai","source":"official","last_updated":"2026-03-23","description":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","context_window":131072,"max_output_tokens":96000,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"reasoning":true},"family":"glm-4.5","model_type":"chat","release_date":"2025-07-25","tagline":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","tools":["function_calling"]},{"id":"z-ai/glm-4.5-air","name":"Z.ai: GLM 4.5 Air","created_by":"z-ai","source":"official","last_updated":"2026-06-18","description":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","context_window":131072,"max_output_tokens":98304,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.13,"output":0.85,"cached_input":0.025},"family":"glm-4.5","model_type":"chat","release_date":"2025-07-25","tagline":"GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.","tools":["function_calling"]},{"id":"z-ai/glm-4.5","name":"Z.ai: GLM 4.5","created_by":"z-ai","source":"official","last_updated":"2026-06-20","description":"GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications.","context_window":131072,"max_output_tokens":98304,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":2.2,"cached_input":0.11},"family":"glm-4.5","model_type":"chat","release_date":"2025-07-25","tagline":"GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications.","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-12-31"},{"id":"z-ai/glm-4.5v","name":"Z.ai: GLM 4.5V","created_by":"z-ai","source":"official","last_updated":"2026-03-24","description":"GLM-4.5V is a vision-language foundation model for multimodal agent applications.","context_window":65536,"max_output_tokens":16384,"modalities":{"input":["text","image"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":1.8,"cached_input":0.11},"family":"glm-4.5","model_type":"chat","release_date":"2025-08-11","tagline":"GLM-4.5V is a vision-language foundation model for multimodal agent applications.","tools":["function_calling"]},{"id":"z-ai/glm-4.6","name":"Z.ai: GLM 4.6","created_by":"z-ai","source":"official","last_updated":"2026-05-14","description":"Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex.","context_window":202752,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.43,"output":1.74,"cached_input":0.08},"family":"glm-4.6","model_type":"chat","release_date":"2025-09-30","tagline":"Compared with GLM-4.5, this generation brings several key improvements:\n\nLonger context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex","tools":["function_calling"],"status":"deprecated","deprecation_date":"2026-05-14"},{"id":"z-ai/glm-4.6v","name":"Z.ai: GLM 4.6V","created_by":"z-ai","source":"official","last_updated":"2026-06-10","description":"GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media.","context_window":131072,"max_output_tokens":32768,"modalities":{"input":["image","text","video"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.3,"output":0.9,"cached_input":0.055},"family":"glm-4.6","model_type":"chat","release_date":"2025-12-08","tagline":"GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media.","tools":["function_calling"]},{"id":"z-ai/glm-4.7-flash","name":"Z.ai: GLM 4.7 Flash","created_by":"z-ai","source":"official","last_updated":"2026-04-25","description":"As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency.","context_window":202752,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.06,"output":0.4,"cached_input":0.01},"family":"glm-4.7","model_type":"chat","release_date":"2026-01-19","tagline":"As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency.","tools":["function_calling"],"max_output_tokens":16384},{"id":"z-ai/glm-4.7","name":"Z.ai: GLM 4.7","created_by":"z-ai","source":"official","last_updated":"2026-05-10","description":"GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution.","context_window":202752,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.4,"output":1.75,"cached_input":0.08},"family":"glm-4.7","model_type":"chat","release_date":"2025-12-22","tagline":"GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution.","tools":["function_calling"]},{"id":"z-ai/glm-5-turbo","name":"Z.ai: GLM 5 Turbo","created_by":"z-ai","source":"official","last_updated":"2026-06-10","description":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios.","context_window":262144,"max_output_tokens":131072,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":1.2,"output":4,"cached_input":0.24},"family":"glm-5","model_type":"chat","release_date":"2026-03-15","tagline":"GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios.","tools":["function_calling"]},{"id":"z-ai/glm-5.1","name":"Z.ai: GLM 5.1","created_by":"z-ai","source":"official","family":"glm-5.1","description":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.","release_date":"2026-04-07","context_window":202752,"model_type":"chat","tagline":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.98,"output":3.08,"cached_input":0.182},"tools":["function_calling"],"last_updated":"2026-06-02","max_output_tokens":131072},{"id":"z-ai/glm-5.2","name":"Z.ai: GLM 5.2","created_by":"z-ai","source":"official","family":"glm-5.2","description":"GLM 5.2 is a large-scale reasoning model from Z.ai.","release_date":"2026-06-16","context_window":1048576,"max_output_tokens":32768,"model_type":"chat","tagline":"GLM 5.2 is a large-scale reasoning model from Z.ai.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.95,"output":3,"cached_input":0.18},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"z-ai/glm-5","name":"Z.ai: GLM 5","created_by":"z-ai","source":"official","last_updated":"2026-06-03","description":"GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows.","context_window":202752,"max_output_tokens":16384,"modalities":{"input":["text"],"output":["text"]},"capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"pricing":{"input":0.6,"output":1.92,"cached_input":0.12},"family":"glm-5","model_type":"chat","release_date":"2026-02-11","tagline":"GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows.","tools":["function_calling"]},{"id":"z-ai/glm-5v-turbo","name":"Z.ai: GLM 5V Turbo","created_by":"z-ai","source":"official","family":"glm-5","description":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks.","release_date":"2026-04-01","context_window":202752,"max_output_tokens":131072,"model_type":"chat","tagline":"GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks.","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true},"modalities":{"input":["image","text","video"],"output":["text"]},"pricing":{"input":1.2,"output":4,"cached_input":0.24},"tools":["function_calling"],"last_updated":"2026-06-29","status":"deprecated","deprecation_date":"2098-12-31"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/perplexity.ts b/packages/npm/src/providers/perplexity.ts
index 123f96c8e..a6f749a61 100644
--- a/packages/npm/src/providers/perplexity.ts
+++ b/packages/npm/src/providers/perplexity.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"perplexity","name":"Perplexity","region":"US","headquarters":"San Francisco, CA","founded":2022,"url":"https://perplexity.ai","api_url":"https://api.perplexity.ai","docs_url":"https://docs.perplexity.ai","pricing_url":"https://docs.perplexity.ai/guides/pricing","description":"AI-powered search engine offering Sonar models with built-in web search and citations.","type":"cloud","openai_compatible":true,"free_tier":false,"models_url":"https://docs.perplexity.ai/guides/model-cards","twitter_url":"https://x.com/perplexity_ai","discord_url":"https://discord.com/invite/perplexity-ai","blog_url":"https://www.perplexity.ai/hub/blog","terms_url":"https://www.perplexity.ai/hub/legal/perplexity-api-terms-of-service","support_url":"https://www.perplexity.ai/hub/helpcenter","icon":"","models":[{"id":"anthropic/claude-haiku-4-5","name":"anthropic/claude-haiku-4-5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":1,"output":5,"cached_input":0.1},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The fastest model with near-frontier intelligence","tagline":"The fastest model with near-frontier intelligence","context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-02","training_data_cutoff":"2025-07"},{"id":"anthropic/claude-opus-4-5","name":"anthropic/claude-opus-4-5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08"},{"id":"anthropic/claude-opus-4-6","name":"anthropic/claude-opus-4-6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The most intelligent broadly available model for agents and coding","tagline":"The most intelligent model for building agents and coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08"},{"id":"anthropic/claude-opus-4-7","name":"anthropic/claude-opus-4-7","created_by":"anthropic","source":"official","family":"claude-opus","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"Our most capable generally available model for complex reasoning and agentic coding","tagline":"Our most capable generally available model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01"},{"id":"anthropic/claude-opus-4-8","name":"anthropic/claude-opus-4-8","created_by":"anthropic","source":"official","family":"claude-opus","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02","description":"Anthropic's most capable Opus-tier model for complex reasoning and agentic coding","tagline":"Anthropic's most capable model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01"},{"id":"anthropic/claude-sonnet-4-5","name":"anthropic/claude-sonnet-4-5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-01","training_data_cutoff":"2025-07"},{"id":"anthropic/claude-sonnet-4-6","name":"anthropic/claude-sonnet-4-6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The best combination of speed and intelligence","tagline":"The best combination of speed and intelligence","context_window":1000000,"max_output_tokens":64000,"knowledge_cutoff":"2025-08","training_data_cutoff":"2026-01"},{"id":"google/gemini-2.5-flash","name":"google/gemini-2.5-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"Our best model in terms of price-performance, offering well-rounded capabilities. 2.5 Flash is best for large scale processing, low-latency, high volume tasks that require thinking, and agentic use cases.","tagline":"Our best model in terms of price-performance, offering well-rounded capabilities.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-2.5-pro","name":"google/gemini-2.5-pro","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","tagline":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3-flash-preview","name":"google/gemini-3-flash-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of state-of-the-art reasoning.","tagline":"The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of stat","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3-pro-preview","name":"google/gemini-3-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.1-flash-lite-preview","name":"google/gemini-3.1-flash-lite-preview","created_by":"google","source":"official","family":"gemini-3.1","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-14","description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 3.1 Flash-Lite is best for high-volume agentic tasks, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.1-flash-lite","name":"google/gemini-3.1-flash-lite","created_by":"google","source":"official","family":"gemini-3.1","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-14","description":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary constraints.","tagline":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary con","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.1-pro-preview","name":"google/gemini-3.1-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.5-flash","name":"google/gemini-3.5-flash","created_by":"google","source":"official","family":"gemini-3.5","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-23","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"nvidia/nemotron-3-super-120b-a12b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.25,"output":2.5},"model_type":"chat","license":"proprietary","parameters":120,"active_parameters":12,"open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5-mini","name":"openai/gpt-5-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 mini is a faster, more cost-efficient version of GPT-5. It's great for","tagline":"Near-frontier intelligence for cost sensitive, low latency, high volume workloads","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05"},{"id":"openai/gpt-5.1","name":"openai/gpt-5.1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1 is our flagship model for coding and agentic tasks with configurable reasoning and non-reasoning effort.","tagline":"The best model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09"},{"id":"openai/gpt-5.2","name":"openai/gpt-5.2","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.4-mini","name":"openai/gpt-5.4-mini","created_by":"openai","source":"official","family":"gpt-5.4","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5,"cached_input":0},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"GPT-5.4 mini brings the strengths of GPT-5.4 to a faster, more efficient","tagline":"Our strongest mini model yet for coding, computer use, and subagents","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.4-nano","name":"openai/gpt-5.4-nano","created_by":"openai","source":"official","family":"gpt-5.4","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25,"cached_input":0},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"GPT-5.4 nano is designed for tasks where speed and cost matter most like","tagline":"Our cheapest GPT-5.4-class model for simple high-volume tasks","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.4","name":"openai/gpt-5.4","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15,"cached_input":0.25},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 is our frontier model for complex professional work.","tagline":"A more affordable model for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.5","name":"openai/gpt-5.5","created_by":"openai","source":"official","family":"gpt-5.5","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"reasoning":true,"batch":true},"pricing":{"input":5,"output":30,"cached_input":0.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01","description":"GPT-5.5 is our newest frontier model for the most complex professional work.","tagline":"A new class of intelligence for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-12"},{"id":"openai/gpt-5","name":"openai/gpt-5","created_by":"openai","source":"official","family":"gpt-5","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-06","pricing":{"input":1.25,"output":10,"cached_input":0.125},"description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09"},{"id":"perplexity/sonar","name":"perplexity/sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.25,"output":2.5,"cached_input":0.0625},"family":"sonar","model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"sonar-deep-research","name":"sonar-deep-research","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":2,"output":8},"model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"A powerful research model capable of conducting exhaustive searches across hundreds of sources, synthesizing expert-level insights, and generating detailed reports with comprehensive analysis.","context_window":128000,"tagline":"Exhaustive research with expert-level insights","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar-deep-research","modalities":{"input":["text"],"output":["text"]}},{"id":"sonar-pro","name":"sonar-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":3,"output":15},"model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"An advanced search model designed for complex queries, delivering deeper content understanding with enhanced search result accuracy and 2x more search results than standard Sonar.","context_window":200000,"tagline":"Advanced search with enhanced search results","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar-pro","modalities":{"input":["text"],"output":["text"]}},{"id":"sonar-reasoning-pro","name":"sonar-reasoning-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":2,"output":8},"model_type":"reasoning","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"A high-performance reasoning model leveraging advanced multi-step Chain-of-Thought (CoT) reasoning and enhanced information retrieval for complex problem-solving.","context_window":128000,"tagline":"Advanced reasoning with enhanced multi-step analysis","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar-reasoning-pro","modalities":{"input":["text"],"output":["text"]}},{"id":"sonar","name":"sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":1,"output":1},"model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"A lightweight, cost-effective search model optimized for quick, grounded answers with real-time web search.","context_window":128000,"tagline":"Fast answers with reliable search results","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar","modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4-1-fast-non-reasoning","name":"xai/grok-4-1-fast-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-4","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"json_mode":true,"batch":true,"vision":true},"pricing":{"input":0.2,"output":0.5,"cached_input":0.05},"model_type":"reasoning","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":2000000,"knowledge_cutoff":"2024-11"},{"id":"xai/grok-4.20-multi-agent","name":"xai/grok-4.20-multi-agent","created_by":"xai","source":"official","family":"grok-4.20","license":"proprietary","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"xai/grok-4.20-non-reasoning","name":"xai/grok-4.20-non-reasoning","created_by":"xai","source":"official","family":"grok-4.20","license":"proprietary","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"xai/grok-4.20-reasoning","name":"xai/grok-4.20-reasoning","created_by":"xai","source":"official","family":"grok-4.20","license":"proprietary","model_type":"reasoning","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"xai/grok-4.3","name":"xai/grok-4.3","created_by":"xai","source":"official","family":"grok-4.3","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"batch":true,"tool_call":true,"structured_output":true,"reasoning":true,"vision":true,"json_mode":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10","context_window":1000000,"knowledge_cutoff":"2024-11"}]};
+export const provider: ProviderWithModels = {"id":"perplexity","name":"Perplexity","region":"US","headquarters":"San Francisco, CA","founded":2022,"url":"https://perplexity.ai","api_url":"https://api.perplexity.ai","docs_url":"https://docs.perplexity.ai","pricing_url":"https://docs.perplexity.ai/guides/pricing","description":"AI-powered search engine offering Sonar models with built-in web search and citations.","type":"cloud","openai_compatible":true,"free_tier":false,"models_url":"https://docs.perplexity.ai/guides/model-cards","twitter_url":"https://x.com/perplexity_ai","discord_url":"https://discord.com/invite/perplexity-ai","blog_url":"https://www.perplexity.ai/hub/blog","terms_url":"https://www.perplexity.ai/hub/legal/perplexity-api-terms-of-service","support_url":"https://www.perplexity.ai/hub/helpcenter","icon":"","models":[{"id":"anthropic/claude-haiku-4-5","name":"anthropic/claude-haiku-4-5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":1,"output":5,"cached_input":0.1},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The fastest model with near-frontier intelligence","tagline":"The fastest model with near-frontier intelligence","context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-02","training_data_cutoff":"2025-07"},{"id":"anthropic/claude-opus-4-5","name":"anthropic/claude-opus-4-5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08"},{"id":"anthropic/claude-opus-4-6","name":"anthropic/claude-opus-4-6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The most intelligent broadly available model for agents and coding","tagline":"The most intelligent model for building agents and coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08"},{"id":"anthropic/claude-opus-4-7","name":"anthropic/claude-opus-4-7","created_by":"anthropic","source":"official","family":"claude-opus","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":5,"output":25,"cached_input":0.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"Our most capable generally available model for complex reasoning and agentic coding","tagline":"Our most capable generally available model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01"},{"id":"anthropic/claude-opus-4-8","name":"anthropic/claude-opus-4-8","created_by":"anthropic","source":"official","family":"claude-opus","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-02","description":"Anthropic's most capable Opus-tier model for complex reasoning and agentic coding","tagline":"Anthropic's most capable model for complex reasoning and agentic coding","context_window":1000000,"max_output_tokens":128000,"knowledge_cutoff":"2026-01","training_data_cutoff":"2026-01"},{"id":"anthropic/claude-sonnet-4-5","name":"anthropic/claude-sonnet-4-5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":200000,"max_output_tokens":64000,"knowledge_cutoff":"2025-01","training_data_cutoff":"2025-07"},{"id":"anthropic/claude-sonnet-4-6","name":"anthropic/claude-sonnet-4-6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15,"cached_input":0.3},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The best combination of speed and intelligence","tagline":"The best combination of speed and intelligence","context_window":1000000,"max_output_tokens":64000,"knowledge_cutoff":"2025-08","training_data_cutoff":"2026-01"},{"id":"google/gemini-2.5-flash","name":"google/gemini-2.5-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Our best model in terms of price-performance, offering well-rounded capabilities. 2.5 Flash is best for large scale processing, low-latency, high volume tasks that require thinking, and agentic use cases.","tagline":"Our best model in terms of price-performance, offering well-rounded capabilities.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-2.5-pro","name":"google/gemini-2.5-pro","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Pro Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","tagline":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3-flash-preview","name":"google/gemini-3-flash-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3 Flash Preview The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of state-of-the-art reasoning.","tagline":"The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of stat","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3-pro-preview","name":"google/gemini-3-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.1-flash-lite-preview","name":"google/gemini-3.1-flash-lite-preview","created_by":"google","source":"official","family":"gemini-3.1","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-14","description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 3.1 Flash-Lite is best for high-volume agentic tasks, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.1-flash-lite","name":"google/gemini-3.1-flash-lite","created_by":"google","source":"official","family":"gemini-3.1","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-14","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash-Lite Gemini 3.1 Flash-Lite is a low-latency, cost-effective multimodal model optimized for high-frequency, lightweight tasks. The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary constraints.","tagline":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary con","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.1-pro-preview","name":"google/gemini-3.1-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Pro Preview Built to refine the performance and reliability of the Gemini 3 Pro series, Gemini 3.1 Pro Preview provides better thinking, improved token efficiency, and a more grounded, factually consistent experience. It's optimized for software engineering behavior and usability, as well as agentic workflows requiring precise tool usage and reliable multi-step execution across real-world domains.","tagline":"The Interactions API is now generally available.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"google/gemini-3.5-flash","name":"google/gemini-3.5-flash","created_by":"google","source":"official","family":"gemini-3.5","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-23","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.5 Flash Gemini 3.5 Flash provides sustained frontier-level intelligence optimized for real-world tasks at a higher speed and lower cost. Designed for the agentic era, it excels at sub-agent deployment, multi-step workflows, and long-horizon tasks at scale. This model is particularly effective for rapid agentic loops involving complex coding cycles and iterations.","tagline":"The Interactions API is now generally available.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01"},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"nvidia/nemotron-3-super-120b-a12b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.25,"output":2.5},"model_type":"chat","license":"proprietary","parameters":120,"active_parameters":12,"open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5-mini","name":"openai/gpt-5-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.25,"output":2,"cached_input":0.025},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 mini is a faster, more cost-efficient version of GPT-5. It's great for","tagline":"Near-frontier intelligence for cost sensitive, low latency, high volume workloads","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05"},{"id":"openai/gpt-5.1","name":"openai/gpt-5.1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10,"cached_input":0.125},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1 is our flagship model for coding and agentic tasks with configurable reasoning and non-reasoning effort.","tagline":"The best model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2024-09"},{"id":"openai/gpt-5.2","name":"openai/gpt-5.2","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14,"cached_input":0.175},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.4-mini","name":"openai/gpt-5.4-mini","created_by":"openai","source":"official","family":"gpt-5.4","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5,"cached_input":0},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"GPT-5.4 mini brings the strengths of GPT-5.4 to a faster, more efficient","tagline":"Our strongest mini model yet for coding, computer use, and subagents","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.4-nano","name":"openai/gpt-5.4-nano","created_by":"openai","source":"official","family":"gpt-5.4","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25,"cached_input":0},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-21","description":"GPT-5.4 nano is designed for tasks where speed and cost matter most like","tagline":"Our cheapest GPT-5.4-class model for simple high-volume tasks","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.4","name":"openai/gpt-5.4","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15,"cached_input":0.25},"model_type":"chat","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 is our frontier model for complex professional work.","tagline":"A more affordable model for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08"},{"id":"openai/gpt-5.5","name":"openai/gpt-5.5","created_by":"openai","source":"official","family":"gpt-5.5","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"reasoning":true,"batch":true},"pricing":{"input":5,"output":30,"cached_input":0.5},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01","description":"GPT-5.5 is our newest frontier model for the most complex professional work.","tagline":"A new class of intelligence for coding and professional work.","context_window":1050000,"max_output_tokens":128000,"knowledge_cutoff":"2025-12"},{"id":"openai/gpt-5","name":"openai/gpt-5","created_by":"openai","source":"official","family":"gpt-5","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-06","pricing":{"input":1.25,"output":10,"cached_input":0.125},"description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09"},{"id":"perplexity/glm-5.2","name":"perplexity/glm-5.2","created_by":"perplexity","source":"official","family":"glm-5.2","license":"proprietary","model_type":"chat","capabilities":{"streaming":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29"},{"id":"perplexity/kimi-k2.7-code","name":"perplexity/kimi-k2.7-code","created_by":"perplexity","source":"official","family":"kimi-k2.7","license":"proprietary","model_type":"chat","capabilities":{"streaming":true},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29"},{"id":"perplexity/sonar","name":"perplexity/sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.25,"output":2.5,"cached_input":0.0625},"family":"sonar","model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"sonar-deep-research","name":"sonar-deep-research","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":2,"output":8},"model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"A powerful research model capable of conducting exhaustive searches across hundreds of sources, synthesizing expert-level insights, and generating detailed reports with comprehensive analysis.","context_window":128000,"tagline":"Exhaustive research with expert-level insights","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar-deep-research","modalities":{"input":["text"],"output":["text"]}},{"id":"sonar-pro","name":"sonar-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":3,"output":15},"model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"An advanced search model designed for complex queries, delivering deeper content understanding with enhanced search result accuracy and 2x more search results than standard Sonar.","context_window":200000,"tagline":"Advanced search with enhanced search results","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar-pro","modalities":{"input":["text"],"output":["text"]}},{"id":"sonar-reasoning-pro","name":"sonar-reasoning-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":2,"output":8},"model_type":"reasoning","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"A high-performance reasoning model leveraging advanced multi-step Chain-of-Thought (CoT) reasoning and enhanced information retrieval for complex problem-solving.","context_window":128000,"tagline":"Advanced reasoning with enhanced multi-step analysis","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar-reasoning-pro","modalities":{"input":["text"],"output":["text"]}},{"id":"sonar","name":"sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","family":"sonar","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":1,"output":1},"model_type":"chat","license":"proprietary","open_weight":false,"tools":["function_calling"],"description":"A lightweight, cost-effective search model optimized for quick, grounded answers with real-time web search.","context_window":128000,"tagline":"Fast answers with reliable search results","page_url":"https://docs.perplexity.ai/docs/sonar/models/sonar","modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4-1-fast-non-reasoning","name":"xai/grok-4-1-fast-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-4","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"json_mode":true,"batch":true,"vision":true},"pricing":{"input":0.2,"output":0.5,"cached_input":0.05},"model_type":"reasoning","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]},"context_window":2000000,"knowledge_cutoff":"2024-11"},{"id":"xai/grok-4.20-multi-agent","name":"xai/grok-4.20-multi-agent","created_by":"xai","source":"official","family":"grok-4.20","license":"proprietary","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"xai/grok-4.20-non-reasoning","name":"xai/grok-4.20-non-reasoning","created_by":"xai","source":"official","family":"grok-4.20","license":"proprietary","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"xai/grok-4.20-reasoning","name":"xai/grok-4.20-reasoning","created_by":"xai","source":"official","family":"grok-4.20","license":"proprietary","model_type":"reasoning","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10"},{"id":"xai/grok-4.3","name":"xai/grok-4.3","created_by":"xai","source":"official","family":"grok-4.3","license":"proprietary","model_type":"chat","capabilities":{"streaming":true,"batch":true,"tool_call":true,"structured_output":true,"reasoning":true,"vision":true,"json_mode":true},"pricing":{"input":1.25,"output":2.5,"cached_input":0.2},"open_weight":false,"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-10","context_window":1000000,"knowledge_cutoff":"2024-11"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/qwen.ts b/packages/npm/src/providers/qwen.ts
index e9f282f72..9a233e03d 100644
--- a/packages/npm/src/providers/qwen.ts
+++ b/packages/npm/src/providers/qwen.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"qwen","name":"Qwen","region":"CN","headquarters":"Hangzhou, Zhejiang","url":"https://qwen.ai","api_url":"https://dashscope-intl.aliyuncs.com/compatible-mode/v1","docs_url":"https://qwen.ai/apiplatform","pricing_url":"https://www.alibabacloud.com/help/en/model-studio/models","description":"Qwen models by Alibaba, available as open-source and via API for text, code, and multimodal tasks.","type":"direct","openai_compatible":true,"free_tier":true,"github_url":"https://github.com/QwenLM","models_url":"https://help.aliyun.com/zh/model-studio/getting-started/models","twitter_url":"https://x.com/alibaba_qwen","discord_url":"https://discord.com/invite/yPEP2vHTu4","blog_url":"https://qwenlm.github.io/blog","terms_url":"https://chat.qwen.ai/legal-agreement/terms-of-service","icon":"","models":[{"id":"cosyvoice-v2","name":"cosyvoice-v2","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3-flash","name":"cosyvoice-v3-flash","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3-plus","name":"cosyvoice-v3-plus","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3.5-flash","name":"cosyvoice-v3.5-flash","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3.5-plus","name":"cosyvoice-v3.5-plus","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"fun-asr-flash-8k-realtime","name":"fun-asr-flash-8k-realtime","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"fun-asr-mtl","name":"fun-asr-mtl","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"fun-asr-realtime","name":"fun-asr-realtime","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"fun-asr","name":"fun-asr","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"paraformer-8k-v2","name":"paraformer-8k-v2","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-8k-v1","name":"paraformer-realtime-8k-v1","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-8k-v2","name":"paraformer-realtime-8k-v2","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-v1","name":"paraformer-realtime-v1","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-v2","name":"paraformer-realtime-v2","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-v2","name":"paraformer-v2","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qvq-max","name":"qvq-max","created_by":"qwen","source":"official","context_window":128000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-coder-plus-1106","name":"qwen-coder-plus-1106","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-coder-plus","name":"qwen-coder-plus","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-coder-turbo-0919","name":"qwen-coder-turbo-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-coder-turbo","name":"qwen-coder-turbo","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-deep-research","name":"qwen-deep-research","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","context_window":1000000,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.007742,"output":0.023367},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen deep research model breaks down complex problems, performs inference and analysis using web search, and generates research reports. [Usage](https://www.alibabacloud.com/help/en/document_detail/2975991.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"The Qwen deep research model can break down complex problems, perform reasoning and analysis using web searches, and generate research reports.[Usage](https://www.alibabacloud.com/help/en/document_detail/2975991.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese Mainland deployment mode](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) is supported.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-doc-turbo","name":"qwen-doc-turbo","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.087,"output":0.144},"endpoints":["chat"],"license":"apache-2.0","description":"You can use the Qwen data mining model to extract structured information from documents for data annotation, content moderation, and other tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2948885.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"The Qwen data mining model extracts structured information from documents for use in data annotation, content moderation, and other applications.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-flash-character","name":"qwen-flash-character","created_by":"qwen","source":"official","family":"qwen","context_window":8000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-flash","name":"qwen-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":1000000,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The fastest and most cost-effective model in the Qwen series, ideal for simple tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-image-2.0-pro","name":"qwen-image-2.0-pro","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen text-to-image model excels at **complex text rendering** , especially for Chinese and English text. [API reference](https://www.alibabacloud.com/help/en/document_detail/2975126.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen text-to-image models excel at rendering complex text, especially bilingual Chinese-English text.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"max_output_tokens":6},{"id":"qwen-image-2.0","name":"qwen-image-2.0","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":6,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-2","name":"qwen-image-2","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-image-edit-max","name":"qwen-image-edit-max","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":6,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-edit-plus","name":"qwen-image-edit-plus","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":6,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-edit","name":"qwen-image-edit","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":1,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-max","name":"qwen-image-max","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":1,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-plus","name":"qwen-image-plus","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":1,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image","name":"qwen-image","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0","max_output_tokens":1,"page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen-long-0125","name":"qwen-long-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-long","name":"qwen-long","created_by":"qwen","source":"official","family":"qwen","context_window":10000000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-math-plus-0816","name":"qwen-math-plus-0816","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-math-plus-0919","name":"qwen-math-plus-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-math-plus","name":"qwen-math-plus","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","context_window":4096,"max_output_tokens":3072,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.574,"output":1.721},"endpoints":["chat"],"license":"proprietary","description":"The Qwen-Math model is a language model designed specifically for solving mathematical problems. [Usage](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen-math-plus?modelGroup=qwen-math-plus) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Qwen-Math is a language model that specializes in solving mathematical problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"qwen-math-turbo-0919","name":"qwen-math-turbo-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-math-turbo","name":"qwen-math-turbo","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.287,"output":0.861},"endpoints":["chat"],"license":"proprietary","description":"The Qwen-Math model is a language model designed specifically for solving mathematical problems. [Usage](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen-math-plus?modelGroup=qwen-math-plus) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Qwen-Math is a language model that specializes in solving mathematical problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"qwen-max-0919","name":"qwen-max-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-max","name":"qwen-max","created_by":"qwen","source":"official","family":"qwen","context_window":32000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"qwen-mt-flash","name":"qwen-mt-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.16,"output":0.49},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":16000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-mt-image","name":"qwen-mt-image","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen image translation models supports translating text in images from 11 languages into Chinese or English. It accurately preserves original layout and content information and offers customizable features such as glossary definition, sensitive words filter, and image entity detection. [API reference](https://www.alibabacloud.com/help/en/document_detail/2977163.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Qwen image translation models support translating text in images from 11 languages into Chinese or English.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-mt-lite","name":"qwen-mt-lite","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.12,"output":0.36},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":16000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-mt-plus","name":"qwen-mt-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":16000,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.46,"output":7.37},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-mt-turbo","name":"qwen-mt-turbo","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.16,"output":0.49},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":16000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-omni-turbo-0119","name":"qwen-omni-turbo-0119","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-omni-turbo-0326","name":"qwen-omni-turbo-0326","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-omni-turbo-realtime","name":"qwen-omni-turbo-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":false,"tools":["function_calling"]},{"id":"qwen-omni-turbo","name":"qwen-omni-turbo","created_by":"qwen","source":"official","family":"qwen","context_window":32000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-plus-0125","name":"qwen-plus-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-0428","name":"qwen-plus-0428","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-0714","name":"qwen-plus-0714","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-0728","name":"qwen-plus-0728","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-character-ja","name":"qwen-plus-character-ja","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":32000,"max_output_tokens":512,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.4},"endpoints":["chat"],"license":"proprietary","description":"Qwen role-playing models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems. Compared with other Qwen models, they improve persona fidelity, topic progression, and empathetic listening.[Usage](https://www.alibabacloud.com/help/en/document_detail/2874763.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen role assumption models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-plus-character","name":"qwen-plus-character","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":32000,"max_output_tokens":4000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.4},"endpoints":["chat"],"license":"proprietary","description":"Qwen role-playing models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems. Compared with other Qwen models, they improve persona fidelity, topic progression, and empathetic listening.[Usage](https://www.alibabacloud.com/help/en/document_detail/2874763.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen role assumption models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-plus-us","name":"qwen-plus-us","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus","name":"qwen-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"proprietary","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-tts-realtime","name":"qwen-tts-realtime","created_by":"qwen","source":"official","family":"qwen","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen-tts","name":"qwen-tts","created_by":"qwen","source":"official","family":"qwen","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen-turbo-0428","name":"qwen-turbo-0428","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-turbo-0715","name":"qwen-turbo-0715","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-turbo-1101","name":"qwen-turbo-1101","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-turbo","name":"qwen-turbo","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"proprietary","description":"Qwen-Turbo is no longer updated. Replace it with Qwen-Flash. Qwen-Flash uses a flexible tiered pricing model for fairer billing. [Usage instructions](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Deep thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Turbo will no longer receive updates.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"context_window":128000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-vl-max-0125","name":"qwen-vl-max-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-0402","name":"qwen-vl-max-0402","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-0408","name":"qwen-vl-max-0408","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-0813","name":"qwen-vl-max-0813","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-1119","name":"qwen-vl-max-1119","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-1230","name":"qwen-vl-max-1230","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-ocr-0413","name":"qwen-vl-ocr-0413","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr-0828","name":"qwen-vl-ocr-0828","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr-1028","name":"qwen-vl-ocr-1028","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr-1120","name":"qwen-vl-ocr-1120","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr","name":"qwen-vl-ocr","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen","context_window":38192,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.07,"output":0.16},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-OCR is a model specifically designed for text extraction. Compared to Qwen-VL models, it focuses more on extracting text from documents, tables, test questions, and handwritten images. It can recognize multiple languages, including English, French, Japanese, Korean, German, Russian, and Italian.[Usage](https://www.alibabacloud.com/help/en/document_detail/2860683.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2996283.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-OCR is a model that specializes in text extraction.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-vl-plus-0102","name":"qwen-vl-plus-0102","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0125","name":"qwen-vl-plus-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0507","name":"qwen-vl-plus-0507","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0710","name":"qwen-vl-plus-0710","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0815","name":"qwen-vl-plus-0815","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-voice-design","name":"qwen-voice-design","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Voice design generates custom voices from text descriptions. It supports multi-language and multi-dimensional voice feature definitions, making it suitable for applications such as ad dubbing, character creation, and audio content production. [Usage](https://www.alibabacloud.com/help/en/document_detail/2938790.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/3000986.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Voice design generates custom voices from text descriptions.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-voice-enrollment","name":"qwen-voice-enrollment","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Voice cloning uses a large model for feature extraction, allowing you to clone voices without training. Provide 10 to 20 seconds of audio to generate a highly similar and natural-sounding custom voice. [Usage](https://www.alibabacloud.com/help/en/document_detail/2938790.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2975034.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Voice cloning uses a large model for feature extraction, allowing you to clone voices without training.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen1.5-110b-chat","name":"qwen1.5-110b-chat","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen1.5","context_window":8000,"max_output_tokens":2000,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":110,"description":"Alibaba Cloud's Qwen1.5 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen1.5 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2-57b-a14b-instruct","name":"qwen2-57b-a14b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen2","context_window":65536,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":57,"active_parameters":14,"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2-72b-instruct","name":"qwen2-72b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen2","context_window":131072,"max_output_tokens":6144,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":72,"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2-7b-instruct","name":"qwen2-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen2","context_window":131072,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2.5-14b-instruct-1m","name":"qwen2.5-14b-instruct-1m","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","context_window":1000000,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.805,"output":3.22},"endpoints":["chat"],"license":"apache-2.0","parameters":14,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-14b-instruct","name":"qwen2.5-14b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.35,"output":1.4},"endpoints":["chat"],"license":"apache-2.0","parameters":14,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-32b-instruct","name":"qwen2.5-32b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.8},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-72b-instruct-1m","name":"qwen2.5-72b-instruct-1m","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":72,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-72b-instruct","name":"qwen2.5-72b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.4,"output":5.6},"endpoints":["chat"],"license":"apache-2.0","parameters":72,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-7b-instruct-1m","name":"qwen2.5-7b-instruct-1m","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.368,"output":1.47},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-7b-instruct","name":"qwen2.5-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.175,"output":0.7},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-math-72b-instruct","name":"qwen2.5-math-72b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwen2.5","context_window":4096,"max_output_tokens":3072,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.574,"output":1.721},"endpoints":["chat"],"license":"apache-2.0","parameters":72,"description":"Qwen-Math is a language model built on the Qwen model to solve math problems. Qwen2.5-Math supports **Chinese** and **English** and integrates multiple inference methods, such as Chain of Thought (CoT), Program of Thought (PoT), and Tool-Integrated Reasoning (TIR). [How to use](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try it online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen2.5-math-72b-instruct) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"A language model built on Qwen, specialized for solving math problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2.5-math-7b-instruct","name":"qwen2.5-math-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.144,"output":0.287},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Qwen-Math is a language model built on the Qwen model to solve math problems. Qwen2.5-Math supports **Chinese** and **English** and integrates multiple inference methods, such as Chain of Thought (CoT), Program of Thought (PoT), and Tool-Integrated Reasoning (TIR). [How to use](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try it online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen2.5-math-72b-instruct) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"A language model built on Qwen, specialized for solving math problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2.5-omni-7b","name":"qwen2.5-omni-7b","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen2.5","context_window":32000,"max_output_tokens":2048,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"A new multimodal understanding and generation LLM trained from Qwen2.5, supporting text, image, audio, and video input understanding. Capable of simultaneous streaming generation of text and speech, with significantly improved multimodal content understanding speed. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"This is a new multimodal understanding and generation model built on Qwen2.5.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-vl-32b-instruct","name":"qwen2.5-vl-32b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":32,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-vl-3b-instruct","name":"qwen2.5-vl-3b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":3,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-vl-72b-instruct","name":"qwen2.5-vl-72b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":72,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-vl-7b-instruct","name":"qwen2.5-vl-7b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":7,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-0.6b","name":"qwen3-0.6b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":0.6,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-1.7b","name":"qwen3-1.7b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":1.7,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-14b","name":"qwen3-14b","created_by":"qwen","source":"official","family":"qwen3","context_window":128000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":14,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"qwen3-235b-a22b-instruct-2507","name":"qwen3-235b-a22b-instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","alias":"qwen3-235b-a22b-instruct","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b-instruct","name":"qwen3-235b-a22b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-235b-a22b-instruct-2507"],"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b-thinking-2507","name":"qwen3-235b-a22b-thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","alias":"qwen3-235b-a22b-thinking","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b-thinking","name":"qwen3-235b-a22b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-235b-a22b-thinking-2507"],"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b","name":"qwen3-235b-a22b","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-instruct-2507","name":"qwen3-30b-a3b-instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","alias":"qwen3-30b-a3b-instruct","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-instruct","name":"qwen3-30b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-30b-a3b-instruct-2507"],"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-thinking-2507","name":"qwen3-30b-a3b-thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","alias":"qwen3-30b-a3b-thinking","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":80000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-thinking","name":"qwen3-30b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-30b-a3b-thinking-2507"],"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":80000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b","name":"qwen3-30b-a3b","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-32b","name":"qwen3-32b","created_by":"qwen","source":"official","family":"qwen3","context_window":128000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":32,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"qwen3-4b","name":"qwen3-4b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":4,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-8b","name":"qwen3-8b","created_by":"qwen","source":"official","family":"qwen3","context_window":128000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":8,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-20"},{"id":"qwen3-asr-flash-filetrans","name":"qwen3-asr-flash-filetrans","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Based on the Qwen multimodal foundation model, this model supports features such as multi-language recognition, singing recognition, and noise rejection. [Usage](https://www.alibabacloud.com/help/en/document_detail/2979031.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2986952.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore. ##### Qwen3-ASR-Flash-Filetrans","tagline":"Based on the Qwen multimodal foundation model, this model supports features such as multi-language recognition, singing recognition, and noise rejection.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-asr-flash-realtime","name":"qwen3-asr-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"audio","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen Real-Time Speech Recognition is a model with automatic language detection. It supports 11 languages and delivers accurate transcription even in complex audio environments. [How to use](https://www.alibabacloud.com/help/en/document_detail/2989727.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2987031.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen Real-Time Speech Recognition is a model with automatic language detection.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-asr-flash","name":"qwen3-asr-flash","created_by":"qwen","source":"official","family":"qwen3","model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3-coder-30b-a3b-instruct","name":"qwen3-coder-30b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-coder-480b-a35b-instruct","name":"qwen3-coder-480b-a35b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","license":"apache-2.0","parameters":480,"active_parameters":35,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-coder-flash","name":"qwen3-coder-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","license":"apache-2.0","context_window":1000000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-coder-next","name":"qwen3-coder-next","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":256000,"max_output_tokens":65536,"model_type":"code","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3-coder-plus","name":"qwen3-coder-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":1000000,"max_output_tokens":65536,"model_type":"code","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen code models: The latest Qwen3-Coder-Plus series are code generation models built on Qwen3, featuring robust Coding Agent capabilities. They excel at tool calling and environment interaction, enabling autonomous programming with exceptional coding skills while retaining general-purpose functionality. [Usage](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/efm/model_experience_center/text?currentTab=textChat&modelId=qwen3-coder-plus) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Coder is a code generation model.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3-livetranslate-flash-realtime","name":"qwen3-livetranslate-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":53248,"max_output_tokens":4096,"model_type":"audio","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-LiveTranslate-Flash-Realtime is a multilingual, real-time audio and video translation model. It can recognize 18 languages and translate them into audio in 10 languages in real time. **Core features:** * **Multi-language support**: Supports 18 languages, such as Chinese, English, French, German, Russian, Japanese, and Korean, and 6 Chinese dialects, including Mandarin, Cantonese, and Sichuanese. * **Visual enhancement**: Uses visual content to improve translation accuracy. The model analyzes lip movements, actions, and on-screen text to improve translation in noisy environments or for words with multiple meanings. * **Low latency**: Achieves simultaneous interpretation latency as low as 3 seconds. * **High-quality simultaneous interpretation**: Addresses cross-language word order issues using semantic unit prediction technology. The real-time translation quality is comparable to offline translation results. * **Natural voice**: Generates natural-sounding, human-like speech. The model adapts its tone and emotion based on the source speech content. [Usage](https://www.alibabacloud.com/help/en/document_detail/2983281.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2983653.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen3-LiveTranslate-Flash-Realtime is a multilingual, real-time audio and video translation model.","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":true},{"id":"qwen3-livetranslate-flash","name":"qwen3-livetranslate-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":53248,"max_output_tokens":4096,"model_type":"translation","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-LiveTranslate-Flash is an audio and video translation model based on the Qwen3-Omni architecture. It supports translation between 18 languages, including Chinese, English, Russian, and French. The model can use visual context to improve translation accuracy and outputs both text and speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2999748.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2999751.html) #### **International** If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen3-LiveTranslate-Flash is an audio and video translation model based on the Qwen3-Omni architecture.","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":true},{"id":"qwen3-max-preview","name":"qwen3-max-preview","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3-max","name":"qwen3-max","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":256000,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true},{"id":"qwen3-next-80b-a3b-instruct","name":"qwen3-next-80b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":80,"active_parameters":3,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-next-80b-a3b-thinking","name":"qwen3-next-80b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-20","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":80,"active_parameters":3,"context_window":128000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-omni-30b-a3b-captioner","name":"qwen3-omni-30b-a3b-captioner","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":65536,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","audio"],"output":["text"]},"pricing":{"input":3.81,"output":3.06},"endpoints":["chat"],"license":"apache-2.0","parameters":30,"active_parameters":3,"description":"Qwen3-Omni-Captioner is an open-source model based on Qwen3-Omni. Without any prompts, it automatically generates accurate, comprehensive descriptions for complex audio, ambient sounds, music, film sound effects, and more. It detects speaker emotions, musical elements (such as genre and instruments), sensitive information, and is suitable for audio content analysis, security review, intent recognition, audio editing, and other fields. [Usage](https://www.alibabacloud.com/help/en/document_detail/2980468.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen3-Omni-Captioner is an open-source model built on Qwen3-Omni.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-omni-flash-0915","name":"qwen3-omni-flash-0915","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen3-omni-flash-realtime","name":"qwen3-omni-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":65536,"max_output_tokens":16384,"model_type":"audio","capabilities":{"streaming":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input with built-in VAD (Voice Activity Detection) to automatically detect the start and end of user speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input and includes built-in Voice Activity Detection (VAD) to automatically detect the start and end of user speech.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-omni-flash","name":"qwen3-omni-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":65536,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Omni accepts multimodal inputs, such as text, images, audio, and video, and generates text or speech responses.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-omni-realtime-flash","name":"qwen3-omni-realtime-flash","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen3-rerank","name":"qwen3-rerank","created_by":"qwen","source":"official","family":"qwen3","model_type":"rerank","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3-tts-flash-realtime","name":"qwen3-tts-flash-realtime","created_by":"qwen","source":"official","family":"qwen3","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen3-tts-flash","name":"qwen3-tts-flash","created_by":"qwen","source":"official","family":"qwen3","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen3-tts-instruct-flash-realtime","name":"qwen3-tts-instruct-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"tts","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"endpoints":["chat"],"license":"apache-2.0","description":"Supports streaming text input and streaming audio output. It can automatically adjust the speech rate based on the text content and punctuation. [Usage](https://www.alibabacloud.com/help/en/document_detail/2938790.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2938849.html) Qwen3-TTS-Instruct-Flash-Realtime supports [Qwen real-time speech synthesis](https://www.alibabacloud.com/help/en/document_detail/2938790.html#12884a10929p9) and can only use the default voice. It does not support cloned or designed voices. Qwen3-TTS-VD-Realtime supports using voices from [Qwen voice design](https://www.alibabacloud.com/help/en/document_detail/3000986.html) for real-time speech synthesis, but does not support the default voice. Qwen3-TTS-VC-Realtime supports using voices from [Qwen voice cloning](https://www.alibabacloud.com/help/en/document_detail/2975034.html) for real-time speech synthesis, but does not support the default voice. Qwen3-TTS-Flash-Realtime and Qwen-TTS-Realtime can only use the default voice. They do not support cloned or designed voices. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore. ##### **Qwen3-TTS-Instruct-Flash-Realtime**","tagline":"Supports streaming text input and streaming audio output.","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true},{"id":"qwen3-tts-instruct-flash","name":"qwen3-tts-instruct-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"tts","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"endpoints":["chat"],"license":"apache-2.0","description":"Supports mixed-language text input and streaming audio output. [Usage](https://www.alibabacloud.com/help/en/document_detail/2879134.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2881635.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore. ##### **Qwen3-TTS-Instruct-Flash**","tagline":"Supports mixed-language text input and streaming audio output.","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true},{"id":"qwen3-vl-235b-a22b-instruct","name":"qwen3-vl-235b-a22b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":235,"active_parameters":22,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-235b-a22b-thinking","name":"qwen3-vl-235b-a22b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":126976,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.4,"output":4},"endpoints":["chat"],"license":"apache-2.0","parameters":235,"active_parameters":22,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-30b-a3b-instruct","name":"qwen3-vl-30b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":30,"active_parameters":3,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-30b-a3b-thinking","name":"qwen3-vl-30b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":126976,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.2,"output":2.4},"endpoints":["chat"],"license":"apache-2.0","parameters":30,"active_parameters":3,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-32b-instruct","name":"qwen3-vl-32b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-32b-thinking","name":"qwen3-vl-32b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":131072,"max_output_tokens":81920,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.16,"output":0.64},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-8b-instruct","name":"qwen3-vl-8b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":8,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-8b-thinking","name":"qwen3-vl-8b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":126976,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.18,"output":2.1},"endpoints":["chat"],"license":"apache-2.0","parameters":8,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-embedding","name":"qwen3-vl-embedding","created_by":"qwen","source":"official","family":"qwen3","model_type":"embed","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":true,"capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3-vl-flash","name":"qwen3-vl-flash","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":81920,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-VL is a text generation model with visual (image) understanding capabilities. It performs OCR (Optical Character Recognition) and further summarizes and infers. For example, it extracts properties from product photos and solves problems based on exercise diagrams. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen-VL models are billed based on the total number of input and output tokens. For rules on calculating image tokens, see [Image and Video Understanding](https://www.alibabacloud.com/help/en/document_detail/2845871.html#7487c7f6eakzl). #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-VL is a text generation model with visual (image) understanding capabilities.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-plus","name":"qwen3-vl-plus","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":262144,"max_output_tokens":81920,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-VL is a text generation model with visual (image) understanding capabilities. It performs OCR (Optical Character Recognition) and further summarizes and infers. For example, it extracts properties from product photos and solves problems based on exercise diagrams. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen-VL models are billed based on the total number of input and output tokens. For rules on calculating image tokens, see [Image and Video Understanding](https://www.alibabacloud.com/help/en/document_detail/2845871.html#7487c7f6eakzl). #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-VL is a text generation model with visual (image) understanding capabilities.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-rerank","name":"qwen3-vl-rerank","created_by":"qwen","source":"official","family":"qwen3","model_type":"rerank","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":true,"capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3.5-122b-a10b","name":"qwen3.5-122b-a10b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":122,"active_parameters":10,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-27b","name":"qwen3.5-27b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":27,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-35b-a3b","name":"qwen3.5-35b-a3b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":35,"active_parameters":3,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-397b-a17b","name":"qwen3.5-397b-a17b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":397,"active_parameters":17,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-flash","name":"qwen3.5-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":1000000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.4},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The fastest and most cost-effective model in the Qwen series, ideal for simple tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-livetranslate-flash-realtime","name":"qwen3.5-livetranslate-flash-realtime","created_by":"qwen","source":"official","family":"qwen3.5","model_type":"audio","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"last_updated":"2026-06-03"},{"id":"qwen3.5-omni-flash-realtime","name":"qwen3.5-omni-flash-realtime","created_by":"qwen","source":"official","family":"qwen3.5","description":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input with built-in VAD (Voice Activity Detection) to automatically detect the start and end of user speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":262144,"max_output_tokens":65536,"model_type":"audio","license":"apache-2.0","tagline":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input and includes built-in VAD (Voice Activity Detection) to automatically detect speech start and end points.[Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client-side events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server-side events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International When the service deployment scope is [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh), data storage is in the **Singapore access region**.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03"},{"id":"qwen3.5-omni-flash","name":"qwen3.5-omni-flash","created_by":"qwen","source":"official","family":"qwen3.5","description":"The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":262144,"model_type":"chat","license":"apache-2.0","tagline":"The Qwen-Omni model accepts multimodal inputs---including text, images, audio, and video---and generates responses in text or speech form.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03"},{"id":"qwen3.5-omni-plus-realtime","name":"qwen3.5-omni-plus-realtime","created_by":"qwen","source":"official","family":"qwen3.5","description":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input with built-in VAD (Voice Activity Detection) to automatically detect the start and end of user speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":262144,"max_output_tokens":65536,"model_type":"audio","license":"apache-2.0","tagline":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input and includes built-in VAD (Voice Activity Detection) to automatically detect speech start and end points.[Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client-side events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server-side events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International When the service deployment scope is [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh), data storage is in the **Singapore access region**.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"vision":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","tools":["function_calling"]},{"id":"qwen3.5-omni-plus","name":"qwen3.5-omni-plus","created_by":"qwen","source":"official","family":"qwen3.5","description":"The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":64000,"model_type":"chat","license":"apache-2.0","tagline":"The Qwen-Omni model accepts multimodal inputs---including text, images, audio, and video---and generates responses in text or speech form.","page_url":"https://www.alibabacloud.com/help/en/model-studio/vision-model","open_weight":true,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat"],"last_updated":"2026-06-03"},{"id":"qwen3.5-plus","name":"qwen3.5-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"proprietary","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"max_output_tokens":64000,"tools":["function_calling"]},{"id":"qwen3.6-27b","name":"qwen3.6-27b","created_by":"qwen","source":"official","family":"qwen3.6","context_window":229376,"model_type":"chat","license":"apache-2.0","parameters":27,"page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-04-24"},{"id":"qwen3.6-35b-a3b","name":"qwen3.6-35b-a3b","created_by":"qwen","source":"official","family":"qwen3.6","context_window":256000,"model_type":"chat","license":"apache-2.0","parameters":35,"active_parameters":3,"page_url":"https://www.alibabacloud.com/help/en/model-studio/vision-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","max_output_tokens":64000,"tools":["function_calling"]},{"id":"qwen3.6-flash","name":"qwen3.6-flash","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":1000000,"model_type":"chat","license":"apache-2.0","tagline":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3.6-max-preview","name":"qwen3.6-max-preview","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":256000,"max_output_tokens":64000,"model_type":"chat","license":"apache-2.0","tagline":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","tools":["function_calling"]},{"id":"qwen3.6-plus","name":"qwen3.6-plus","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":1000000,"model_type":"chat","license":"apache-2.0","tagline":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","max_output_tokens":64000,"tools":["function_calling"]},{"id":"qwen3.7-max-preview","name":"qwen3.7-max-preview","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"max_output_tokens":64000,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3.7-max","name":"qwen3.7-max","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3.7-plus","name":"qwen3.7-plus","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3","name":"qwen3","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwq-32b-preview","name":"qwq-32b-preview","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwq","context_window":32768,"max_output_tokens":16384,"model_type":"reasoning","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.287,"output":0.861},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"qwq-32b-preview is an experimental research model developed by the Qwen team in 2024, focused on enhancing AI reasoning capabilities, especially in math and programming. See [QwQ official blog](https://qwenlm.github.io/zh/blog/qwq-32b-preview/#%E6%A8%A1%E5%9E%8B%E5%B1%80%E9%99%90%E6%80%A7) for model limitations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwq-32b-preview) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"qwq-32b-preview is an experimental research model developed by the Qwen team in 2024.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwq-32b","name":"qwq-32b","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwq","context_window":131072,"max_output_tokens":32768,"model_type":"reasoning","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.287,"output":0.861},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"QwQ is a reasoning model trained from Qwen2.5-32B, with significantly enhanced reasoning capabilities through reinforcement learning. Its core metrics (AIME 24/25, LiveCodeBench) and some general metrics (IFEval, LiveBench, etc.) match DeepSeek-R1 full version levels and significantly outperform DeepSeek-R1-Distill-Qwen-32B, which is also based on Qwen2.5-32B. [Usage](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"QwQ is an inference model trained on Qwen2.5-32B.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwq-plus","name":"qwq-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwq","context_window":128000,"max_output_tokens":32768,"model_type":"reasoning","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The QwQ reasoning model, trained on the Qwen2.5 model, uses reinforcement learning to significantly improve its model inference capabilities. The model's core metrics for math and code (AIME 24/25, LiveCodeBench) and general metrics (IFEval, LiveBench) are on par with the full-performance version of DeepSeek-R1. [Usage](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"QwQ is a reasoning model trained on the Qwen2.5 base and significantly enhanced through reinforcement learning.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"pricing":{"input":0.8,"output":2.4},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"text-embedding-v3","name":"text-embedding-v3","created_by":"qwen","source":"official","family":"text-embedding","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"text-embedding-v4","name":"text-embedding-v4","created_by":"qwen","source":"official","family":"text-embedding","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"tongyi-embedding-vision-flash","name":"tongyi-embedding-vision-flash","created_by":"qwen","source":"official","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"tongyi-embedding-vision-plus","name":"tongyi-embedding-vision-plus","created_by":"qwen","source":"official","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"wan-pro","name":"wan-pro","created_by":"qwen","source":"official","last_updated":"2026-04-21","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan","model_type":"video","license":"proprietary","description":"Offers standard and professional service modes. Uses a portrait image and reference video to transfer the video subject's actions and expressions to the portrait image, generating a dynamic action video. [API reference](https://www.alibabacloud.com/help/en/document_detail/2981852.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"This service offers Standard and Professional modes.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.1-kf2v-plus","name":"wan2.1-kf2v-plus","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.1","license":"proprietary","description":"The Wan first-and-last-frame image-to-video model generates smooth, fluid videos using just two images---the first and last frames---plus your prompt. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2880649.html) \\| [Try it now](https://wan.video/wanxiang/videoCreation) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"Wan first-and-last-frame video generation models generate smooth, fluid videos from just two input frames, the first and last, plus a prompt.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.1-t2i-plus","name":"wan2.1-t2i-plus","created_by":"qwen","source":"official","family":"wan2.1","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.1-t2i-turbo","name":"wan2.1-t2i-turbo","created_by":"qwen","source":"official","family":"wan2.1","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.1-vace-plus","name":"wan2.1-vace-plus","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.1","license":"proprietary","description":"The Wan video editing model supports multimodal inputs, including text, images, and video, enabling video generation and general editing tasks. [API reference (Video Editing 2.7)](https://www.alibabacloud.com/help/en/document_detail/3021842.html) \\| [API reference (Video Editing 2.1)](https://www.alibabacloud.com/help/en/document_detail/2922183.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"Wan general video editing models support multimodal inputs, including text, images, and video, and perform both video generation and general editing tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-animate-mix","name":"wan2.2-animate-mix","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Offers standard and professional service modes. Uses a portrait image and reference video to replace the main subject in the video with the portrait image, while preserving the original video's scene, lighting, and hue. [API reference](https://www.alibabacloud.com/help/en/document_detail/2982219.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"This service offers Standard and Professional modes.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-animate-move","name":"wan2.2-animate-move","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Offers standard and professional service modes. Uses a portrait image and reference video to transfer the video subject's actions and expressions to the portrait image, generating a dynamic action video. [API reference](https://www.alibabacloud.com/help/en/document_detail/2981852.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"This service offers Standard and Professional modes.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-kf2v-flash","name":"wan2.2-kf2v-flash","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"The Wan first-and-last-frame image-to-video model generates smooth, fluid videos using just two images---the first and last frames---plus your prompt. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2880649.html) \\| [Try it now](https://wan.video/wanxiang/videoCreation) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"Wan first-and-last-frame video generation models generate smooth, fluid videos from just two input frames, the first and last, plus a prompt.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-s2v-detect","name":"wan2.2-s2v-detect","created_by":"qwen","source":"official","last_updated":"2026-04-03","model_type":"video","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Generate natural talking, singing, or performing videos from a single portrait image and audio. Call the following models in order. [wan2.2-s2v image detection](https://www.alibabacloud.com/help/en/document_detail/2978214.html) \\| [wan2.2-s2v video generation](https://www.alibabacloud.com/help/en/document_detail/2978215.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"You can generate natural-speaking, singing, or performing videos from a single portrait image and audio file.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-s2v","name":"wan2.2-s2v","created_by":"qwen","source":"official","last_updated":"2026-04-03","model_type":"video","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Generate natural talking, singing, or performing videos from a single portrait image and audio. Call the following models in order. [wan2.2-s2v image detection](https://www.alibabacloud.com/help/en/document_detail/2978214.html) \\| [wan2.2-s2v video generation](https://www.alibabacloud.com/help/en/document_detail/2978215.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"You can generate natural-speaking, singing, or performing videos from a single portrait image and audio file.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-t2i-flash","name":"wan2.2-t2i-flash","created_by":"qwen","source":"official","family":"wan2.2","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.2-t2i-plus","name":"wan2.2-t2i-plus","created_by":"qwen","source":"official","family":"wan2.2","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.5-i2i-preview","name":"wan2.5-i2i-preview","created_by":"qwen","source":"official","last_updated":"2026-06-03","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["video"]},"endpoints":["chat"],"family":"wan2.5","model_type":"video","license":"proprietary","description":"The Wan general image editing 2.5 model supports inputting text, a single image, or multiple images to perform subject-consistent image editing and multi-image fusion creation. [API reference](https://www.alibabacloud.com/help/en/document_detail/2982258.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Wan general image editing 2.5 supports subject-consistent editing using text, a single image, or multiple images.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"wan2.5-t2i-preview","name":"wan2.5-t2i-preview","created_by":"qwen","source":"official","family":"wan2.5","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.6-i2v","name":"wan2.6-i2v","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan image-to-video model uses your input image as the first frame, then generates a video based on your prompt. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2867393.html) \\| [Try it now](https://modelstudio.console.alibabacloud.com/?tab=dashboard#/efm/model_experience_center/vision?currentTab=videoGenerate) ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan image-to-video models use your input image as the first frame, then generate a video based on your prompt, delivering rich artistic styles and cinematic quality.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.6-image","name":"wan2.6-image","created_by":"qwen","source":"official","last_updated":"2026-06-03","model_type":"image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan image generation model supports text-to-image, text-to-series, image-to-series, image editing, and multi-image reference generation, meeting diverse generation and integration needs. [API reference - 2.7](https://www.alibabacloud.com/help/en/document_detail/3026980.html) \\| [API reference - 2.6](https://www.alibabacloud.com/help/en/document_detail/3001143.html) Billing rule: You are charged for the number of **successfully generated images**. Failed attempts are not billed and do not consume your free quota. ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan image generation models support image editing and mixed text-image outputs to meet diverse generation and integration needs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"wan2.6-r2v","name":"wan2.6-r2v","created_by":"qwen","source":"official","last_updated":"2026-04-22","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text","audio","video"],"output":["text"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan reference-to-video model lets you generate performance videos using characters and voices from reference videos or images. [API reference](https://www.alibabacloud.com/help/en/document_detail/3001146.html) Billing rule: Both input and output videos are billed by **video duration in seconds**. Failed requests are not billed and do not consume your free quota. * Input video duration is capped at **5 seconds** . See [Wan reference-to-video](https://www.alibabacloud.com/help/en/document_detail/3001146.html#f79461ca408qn) for details. * Output video duration equals the **duration of successfully generated video**. ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"The Wan reference-to-video model generates performance videos by referencing a character's appearance from an input video or image, the voice from the input video, and text prompts.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.6-t2i","name":"wan2.6-t2i","created_by":"qwen","source":"official","last_updated":"2026-06-03","model_type":"image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan text-to-image model generates high-quality images from text. [API reference](https://www.alibabacloud.com/help/en/document_detail/2862677.html) \\| [Try online](https://wan.video/wanxiang/creation) ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan text-to-image models generate high-quality images from simple text prompts.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"wan2.6-t2v","name":"wan2.6-t2v","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan text-to-video model generates videos from a single sentence. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2865250.html) \\| [Try it now](https://modelstudio.console.alibabacloud.com/?tab=dashboard#/efm/model_experience_center/vision?currentTab=videoGenerate) ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan text-to-video models generate videos from a single sentence, delivering rich artistic styles and cinematic quality.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.7-i2v","name":"wan2.7-i2v","created_by":"qwen","source":"official","family":"wan2.7","description":"The Wan image-to-video model has been upgraded to support multimodal inputs (text, image, audio, and video), and can perform three types of tasks: **first-frame-to-video, first-and-last-frame-to-video, and video continuation** . [API reference](https://www.alibabacloud.com/help/en/document_detail/3025059.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","model_type":"video","license":"proprietary","tagline":"The Wan image-to-video model has been upgraded to support multimodal inputs (text, image, audio, and video), and can perform three types of tasks: **first-frame-to-video, first-and-last-frame-to-video, and video continuation** .","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text","video"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-04-21"},{"id":"wan2.7-image-pro","name":"wan2.7-image-pro","created_by":"qwen","source":"official","family":"wan2.7","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.7-image","name":"wan2.7-image","created_by":"qwen","source":"official","family":"wan2.7","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.7-videoedit","name":"wan2.7-videoedit","created_by":"qwen","source":"official","family":"wan2.7","description":"The Wan video editing model supports multimodal inputs, including text, images, and video, enabling video generation and general editing tasks. [API reference (Video Editing 2.7)](https://www.alibabacloud.com/help/en/document_detail/3021842.html) \\| [API reference (Video Editing 2.1)](https://www.alibabacloud.com/help/en/document_detail/2922183.html) ## International When the service deployment scope is[Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh), data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","model_type":"video","license":"proprietary","tagline":"The Wan video editing model supports multimodal inputs, including text, images, and video, enabling video generation and general editing tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text","video"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-04-03"},{"id":"wanx2.1-imageedit","name":"wanx2.1-imageedit","created_by":"qwen","source":"official","last_updated":"2026-06-03","model_type":"image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["chat"],"family":"wan","license":"proprietary","description":"The Wan general image editing model supports diverse image editing tasks using simple instructions. Use it for image outpainting, watermark removal, style transfer, image inpainting, and image enhancement. [Usage](https://www.alibabacloud.com/help/en/document_detail/2874281.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2868981.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Wan general image editing models enable diverse image editing tasks using simple instructions.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"z-image-turbo","name":"z-image-turbo","created_by":"qwen","source":"official","max_output_tokens":1,"model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"}]};
+export const provider: ProviderWithModels = {"id":"qwen","name":"Qwen","region":"CN","headquarters":"Hangzhou, Zhejiang","url":"https://qwen.ai","api_url":"https://dashscope-intl.aliyuncs.com/compatible-mode/v1","docs_url":"https://qwen.ai/apiplatform","pricing_url":"https://www.alibabacloud.com/help/en/model-studio/models","description":"Qwen models by Alibaba, available as open-source and via API for text, code, and multimodal tasks.","type":"direct","openai_compatible":true,"free_tier":true,"github_url":"https://github.com/QwenLM","models_url":"https://help.aliyun.com/zh/model-studio/getting-started/models","twitter_url":"https://x.com/alibaba_qwen","discord_url":"https://discord.com/invite/yPEP2vHTu4","blog_url":"https://qwenlm.github.io/blog","terms_url":"https://chat.qwen.ai/legal-agreement/terms-of-service","icon":"","models":[{"id":"cosyvoice-v2","name":"cosyvoice-v2","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3-flash","name":"cosyvoice-v3-flash","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3-plus","name":"cosyvoice-v3-plus","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3.5-flash","name":"cosyvoice-v3.5-flash","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"cosyvoice-v3.5-plus","name":"cosyvoice-v3.5-plus","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"fun-asr-flash-8k-realtime","name":"fun-asr-flash-8k-realtime","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"fun-asr-mtl","name":"fun-asr-mtl","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"fun-asr-realtime","name":"fun-asr-realtime","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"fun-asr","name":"fun-asr","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"paraformer-8k-v2","name":"paraformer-8k-v2","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-8k-v1","name":"paraformer-realtime-8k-v1","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-8k-v2","name":"paraformer-realtime-8k-v2","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-v1","name":"paraformer-realtime-v1","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-realtime-v2","name":"paraformer-realtime-v2","created_by":"qwen","source":"official","model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"last_updated":"2026-06-03"},{"id":"paraformer-v2","name":"paraformer-v2","created_by":"qwen","source":"official","model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qvq-max","name":"qvq-max","created_by":"qwen","source":"official","context_window":128000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-coder-plus-1106","name":"qwen-coder-plus-1106","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-coder-plus","name":"qwen-coder-plus","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-coder-turbo-0919","name":"qwen-coder-turbo-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-coder-turbo","name":"qwen-coder-turbo","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"code","license":"apache-2.0"},{"id":"qwen-deep-research","name":"qwen-deep-research","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","context_window":1000000,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.007742,"output":0.023367},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen deep research model breaks down complex problems, performs inference and analysis using web search, and generates research reports. [Usage](https://www.alibabacloud.com/help/en/document_detail/2975991.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"The Qwen deep research model can break down complex problems, perform reasoning and analysis using web searches, and generate research reports.[Usage](https://www.alibabacloud.com/help/en/document_detail/2975991.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese Mainland deployment mode](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) is supported.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-doc-turbo","name":"qwen-doc-turbo","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","context_window":262144,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.087,"output":0.144},"endpoints":["chat"],"license":"apache-2.0","description":"You can use the Qwen data mining model to extract structured information from documents for data annotation, content moderation, and other tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2948885.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"The Qwen data mining model extracts structured information from documents for use in data annotation, content moderation, and other applications.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-flash-character","name":"qwen-flash-character","created_by":"qwen","source":"official","family":"qwen","context_window":8000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-flash","name":"qwen-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":1000000,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The fastest and most cost-effective model in the Qwen series, ideal for simple tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-image-2.0-pro","name":"qwen-image-2.0-pro","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen text-to-image model excels at **complex text rendering** , especially for Chinese and English text. [API reference](https://www.alibabacloud.com/help/en/document_detail/2975126.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen text-to-image models excel at rendering complex text, especially bilingual Chinese-English text.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"max_output_tokens":6},{"id":"qwen-image-2.0","name":"qwen-image-2.0","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":6,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-2","name":"qwen-image-2","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-image-edit-max","name":"qwen-image-edit-max","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":6,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-edit-plus","name":"qwen-image-edit-plus","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":6,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-edit","name":"qwen-image-edit","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":1,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-max","name":"qwen-image-max","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":1,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image-plus","name":"qwen-image-plus","created_by":"qwen","source":"official","family":"qwen","max_output_tokens":1,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen-image","name":"qwen-image","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0","max_output_tokens":1,"page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen-long-0125","name":"qwen-long-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-long","name":"qwen-long","created_by":"qwen","source":"official","family":"qwen","context_window":10000000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-math-plus-0816","name":"qwen-math-plus-0816","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-math-plus-0919","name":"qwen-math-plus-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-math-plus","name":"qwen-math-plus","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","context_window":4096,"max_output_tokens":3072,"model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.574,"output":1.721},"endpoints":["chat"],"license":"proprietary","description":"The Qwen-Math model is a language model designed specifically for solving mathematical problems. [Usage](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen-math-plus?modelGroup=qwen-math-plus) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Qwen-Math is a language model that specializes in solving mathematical problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"qwen-math-turbo-0919","name":"qwen-math-turbo-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-math-turbo","name":"qwen-math-turbo","created_by":"qwen","source":"official","last_updated":"2026-04-08","family":"qwen","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.287,"output":0.861},"endpoints":["chat"],"license":"proprietary","description":"The Qwen-Math model is a language model designed specifically for solving mathematical problems. [Usage](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen-math-plus?modelGroup=qwen-math-plus) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Qwen-Math is a language model that specializes in solving mathematical problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"qwen-max-0919","name":"qwen-max-0919","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-max","name":"qwen-max","created_by":"qwen","source":"official","family":"qwen","context_window":128000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"qwen-mt-flash","name":"qwen-mt-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.16,"output":0.49},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":16000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-mt-image","name":"qwen-mt-image","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen image translation models supports translating text in images from 11 languages into Chinese or English. It accurately preserves original layout and content information and offers customizable features such as glossary definition, sensitive words filter, and image entity detection. [API reference](https://www.alibabacloud.com/help/en/document_detail/2977163.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Qwen image translation models support translating text in images from 11 languages into Chinese or English.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-mt-lite","name":"qwen-mt-lite","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.12,"output":0.36},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":16000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-mt-plus","name":"qwen-mt-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":16000,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":2.46,"output":7.37},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-mt-turbo","name":"qwen-mt-turbo","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.16,"output":0.49},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-MT is a flagship Large Language Model (LLM) for translation and a comprehensive upgrade of Qwen 3. It supports translation between 92 languages, including Chinese, English, Japanese, Korean, French, Spanish, German, Thai, Indonesian, Vietnamese, and Arabic. The model's performance and translation quality are significantly improved. It provides enhanced stability for terminology customization, format preservation, and domain-specific prompting, resulting in more accurate and natural translations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2860790.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-MT is a flagship Large Language Model (LLM) for translation, fully upgraded from Qwen 3.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":16000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-omni-turbo-0119","name":"qwen-omni-turbo-0119","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-omni-turbo-0326","name":"qwen-omni-turbo-0326","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-omni-turbo-realtime","name":"qwen-omni-turbo-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"vision":true},"model_type":"audio","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":false,"tools":["function_calling"]},{"id":"qwen-omni-turbo","name":"qwen-omni-turbo","created_by":"qwen","source":"official","family":"qwen","context_window":32000,"model_type":"chat","reasoning_tokens":true,"license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen-plus-0125","name":"qwen-plus-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-0428","name":"qwen-plus-0428","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-0714","name":"qwen-plus-0714","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-0728","name":"qwen-plus-0728","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus-character-ja","name":"qwen-plus-character-ja","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":32000,"max_output_tokens":512,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.4},"endpoints":["chat"],"license":"proprietary","description":"Qwen role-playing models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems. Compared with other Qwen models, they improve persona fidelity, topic progression, and empathetic listening.[Usage](https://www.alibabacloud.com/help/en/document_detail/2874763.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen role assumption models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-plus-character","name":"qwen-plus-character","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","context_window":32000,"max_output_tokens":4000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.5,"output":1.4},"endpoints":["chat"],"license":"proprietary","description":"Qwen role-playing models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems. Compared with other Qwen models, they improve persona fidelity, topic progression, and empathetic listening.[Usage](https://www.alibabacloud.com/help/en/document_detail/2874763.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen role assumption models are designed for anthropomorphic dialog scenarios, such as virtual socializing, game NPCs, IP replication, and hardware, toys, or in-vehicle systems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-plus-us","name":"qwen-plus-us","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-plus","name":"qwen-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"proprietary","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-tts-realtime","name":"qwen-tts-realtime","created_by":"qwen","source":"official","family":"qwen","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen-tts","name":"qwen-tts","created_by":"qwen","source":"official","family":"qwen","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen-turbo-0428","name":"qwen-turbo-0428","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-turbo-0715","name":"qwen-turbo-0715","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-turbo-1101","name":"qwen-turbo-1101","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-turbo","name":"qwen-turbo","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"proprietary","description":"Qwen-Turbo is no longer updated. Replace it with Qwen-Flash. Qwen-Flash uses a flexible tiered pricing model for fairer billing. [Usage instructions](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Deep thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Turbo will no longer receive updates.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen-vl-max-0125","name":"qwen-vl-max-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-0402","name":"qwen-vl-max-0402","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-0408","name":"qwen-vl-max-0408","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-0813","name":"qwen-vl-max-0813","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-1119","name":"qwen-vl-max-1119","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-max-1230","name":"qwen-vl-max-1230","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-ocr-0413","name":"qwen-vl-ocr-0413","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr-0828","name":"qwen-vl-ocr-0828","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr-1028","name":"qwen-vl-ocr-1028","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr-1120","name":"qwen-vl-ocr-1120","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen-vl-ocr","name":"qwen-vl-ocr","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen","context_window":38192,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.07,"output":0.16},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-OCR is a model specifically designed for text extraction. Compared to Qwen-VL models, it focuses more on extracting text from documents, tables, test questions, and handwritten images. It can recognize multiple languages, including English, French, Japanese, Korean, German, Russian, and Italian.[Usage](https://www.alibabacloud.com/help/en/document_detail/2860683.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2996283.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-OCR is a model that specializes in text extraction.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-vl-plus-0102","name":"qwen-vl-plus-0102","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0125","name":"qwen-vl-plus-0125","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0507","name":"qwen-vl-plus-0507","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0710","name":"qwen-vl-plus-0710","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-vl-plus-0815","name":"qwen-vl-plus-0815","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen","capabilities":{"streaming":true,"tool_call":true},"model_type":"chat","license":"proprietary"},{"id":"qwen-voice-design","name":"qwen-voice-design","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Voice design generates custom voices from text descriptions. It supports multi-language and multi-dimensional voice feature definitions, making it suitable for applications such as ad dubbing, character creation, and audio content production. [Usage](https://www.alibabacloud.com/help/en/document_detail/2938790.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/3000986.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Voice design generates custom voices from text descriptions.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen-voice-enrollment","name":"qwen-voice-enrollment","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Voice cloning uses a large model for feature extraction, allowing you to clone voices without training. Provide 10 to 20 seconds of audio to generate a highly similar and natural-sounding custom voice. [Usage](https://www.alibabacloud.com/help/en/document_detail/2938790.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2975034.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Voice cloning uses a large model for feature extraction, allowing you to clone voices without training.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen1.5-110b-chat","name":"qwen1.5-110b-chat","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen1.5","context_window":8000,"max_output_tokens":2000,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":110,"description":"Alibaba Cloud's Qwen1.5 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen1.5 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2-57b-a14b-instruct","name":"qwen2-57b-a14b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen2","context_window":65536,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":57,"active_parameters":14,"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2-72b-instruct","name":"qwen2-72b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen2","context_window":131072,"max_output_tokens":6144,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":72,"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2-7b-instruct","name":"qwen2-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-01","family":"qwen2","context_window":131072,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Alibaba Cloud's Qwen2 open-source version. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt)","tagline":"The Qwen2 open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2.5-14b-instruct-1m","name":"qwen2.5-14b-instruct-1m","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","context_window":1000000,"max_output_tokens":8192,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.805,"output":3.22},"endpoints":["chat"],"license":"apache-2.0","parameters":14,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-14b-instruct","name":"qwen2.5-14b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.35,"output":1.4},"endpoints":["chat"],"license":"apache-2.0","parameters":14,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-32b-instruct","name":"qwen2.5-32b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.7,"output":2.8},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-72b-instruct-1m","name":"qwen2.5-72b-instruct-1m","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":72,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-72b-instruct","name":"qwen2.5-72b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":1.4,"output":5.6},"endpoints":["chat"],"license":"apache-2.0","parameters":72,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-7b-instruct-1m","name":"qwen2.5-7b-instruct-1m","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.368,"output":1.47},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-7b-instruct","name":"qwen2.5-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.175,"output":0.7},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen2.5 is a large language model series from Qwen.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"context_window":1000000,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-math-72b-instruct","name":"qwen2.5-math-72b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwen2.5","context_window":4096,"max_output_tokens":3072,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.574,"output":1.721},"endpoints":["chat"],"license":"apache-2.0","parameters":72,"description":"Qwen-Math is a language model built on the Qwen model to solve math problems. Qwen2.5-Math supports **Chinese** and **English** and integrates multiple inference methods, such as Chain of Thought (CoT), Program of Thought (PoT), and Tool-Integrated Reasoning (TIR). [How to use](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try it online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen2.5-math-72b-instruct) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"A language model built on Qwen, specialized for solving math problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2.5-math-7b-instruct","name":"qwen2.5-math-7b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwen2.5","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.144,"output":0.287},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"Qwen-Math is a language model built on the Qwen model to solve math problems. Qwen2.5-Math supports **Chinese** and **English** and integrates multiple inference methods, such as Chain of Thought (CoT), Program of Thought (PoT), and Tool-Integrated Reasoning (TIR). [How to use](https://www.alibabacloud.com/help/en/document_detail/2849934.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try it online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwen2.5-math-72b-instruct) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"A language model built on Qwen, specialized for solving math problems.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen2.5-omni-7b","name":"qwen2.5-omni-7b","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen2.5","context_window":1000000,"max_output_tokens":2048,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":7,"description":"A new multimodal understanding and generation LLM trained from Qwen2.5, supporting text, image, audio, and video input understanding. Capable of simultaneous streaming generation of text and speech, with significantly improved multimodal content understanding speed. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"This is a new multimodal understanding and generation model built on Qwen2.5.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen2.5-vl-32b-instruct","name":"qwen2.5-vl-32b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":32,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-vl-3b-instruct","name":"qwen2.5-vl-3b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":3,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-vl-72b-instruct","name":"qwen2.5-vl-72b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":72,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen2.5-vl-7b-instruct","name":"qwen2.5-vl-7b-instruct","created_by":"qwen","source":"official","family":"qwen2.5","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":7,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"vision":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-0.6b","name":"qwen3-0.6b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":0.6,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-1.7b","name":"qwen3-1.7b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":1.7,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-14b","name":"qwen3-14b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":14,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"qwen3-235b-a22b-instruct-2507","name":"qwen3-235b-a22b-instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","alias":"qwen3-235b-a22b-instruct","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b-instruct","name":"qwen3-235b-a22b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-235b-a22b-instruct-2507"],"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b-thinking-2507","name":"qwen3-235b-a22b-thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","alias":"qwen3-235b-a22b-thinking","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b-thinking","name":"qwen3-235b-a22b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-235b-a22b-thinking-2507"],"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-235b-a22b","name":"qwen3-235b-a22b","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":235,"active_parameters":22,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-instruct-2507","name":"qwen3-30b-a3b-instruct-2507","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","alias":"qwen3-30b-a3b-instruct","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-instruct","name":"qwen3-30b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-30b-a3b-instruct-2507"],"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-thinking-2507","name":"qwen3-30b-a3b-thinking-2507","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","alias":"qwen3-30b-a3b-thinking","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b-thinking","name":"qwen3-30b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"snapshots":["qwen3-30b-a3b-thinking-2507"],"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-30b-a3b","name":"qwen3-30b-a3b","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-32b","name":"qwen3-32b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":32,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"qwen3-4b","name":"qwen3-4b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":4,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3-8b","name":"qwen3-8b","created_by":"qwen","source":"official","family":"qwen3","context_window":256000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","parameters":8,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"qwen3-asr-flash-filetrans","name":"qwen3-asr-flash-filetrans","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Based on the Qwen multimodal foundation model, this model supports features such as multi-language recognition, singing recognition, and noise rejection. [Usage](https://www.alibabacloud.com/help/en/document_detail/2979031.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2986952.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore. ##### Qwen3-ASR-Flash-Filetrans","tagline":"Based on the Qwen multimodal foundation model, this model supports features such as multi-language recognition, singing recognition, and noise rejection.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-asr-flash-realtime","name":"qwen3-asr-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"audio","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen Real-Time Speech Recognition is a model with automatic language detection. It supports 11 languages and delivers accurate transcription even in complex audio environments. [How to use](https://www.alibabacloud.com/help/en/document_detail/2989727.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2987031.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen Real-Time Speech Recognition is a model with automatic language detection.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-asr-flash","name":"qwen3-asr-flash","created_by":"qwen","source":"official","family":"qwen3","model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3-coder-30b-a3b-instruct","name":"qwen3-coder-30b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","license":"apache-2.0","parameters":30,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-coder-480b-a35b-instruct","name":"qwen3-coder-480b-a35b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","license":"apache-2.0","parameters":480,"active_parameters":35,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-coder-flash","name":"qwen3-coder-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"code","license":"apache-2.0","context_window":1000000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-coder-next","name":"qwen3-coder-next","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":256000,"max_output_tokens":65536,"model_type":"code","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3-coder-plus","name":"qwen3-coder-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":1000000,"max_output_tokens":65536,"model_type":"code","capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen code models: The latest Qwen3-Coder-Plus series are code generation models built on Qwen3, featuring robust Coding Agent capabilities. They excel at tool calling and environment interaction, enabling autonomous programming with exceptional coding skills while retaining general-purpose functionality. [Usage](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/efm/model_experience_center/text?currentTab=textChat&modelId=qwen3-coder-plus) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Coder is a code generation model.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3-livetranslate-flash-realtime","name":"qwen3-livetranslate-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":53248,"max_output_tokens":4096,"model_type":"audio","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-LiveTranslate-Flash-Realtime is a multilingual, real-time audio and video translation model. It can recognize 18 languages and translate them into audio in 10 languages in real time. **Core features:** * **Multi-language support**: Supports 18 languages, such as Chinese, English, French, German, Russian, Japanese, and Korean, and 6 Chinese dialects, including Mandarin, Cantonese, and Sichuanese. * **Visual enhancement**: Uses visual content to improve translation accuracy. The model analyzes lip movements, actions, and on-screen text to improve translation in noisy environments or for words with multiple meanings. * **Low latency**: Achieves simultaneous interpretation latency as low as 3 seconds. * **High-quality simultaneous interpretation**: Addresses cross-language word order issues using semantic unit prediction technology. The real-time translation quality is comparable to offline translation results. * **Natural voice**: Generates natural-sounding, human-like speech. The model adapts its tone and emotion based on the source speech content. [Usage](https://www.alibabacloud.com/help/en/document_detail/2983281.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2983653.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen3-LiveTranslate-Flash-Realtime is a multilingual, real-time audio and video translation model.","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":true},{"id":"qwen3-livetranslate-flash","name":"qwen3-livetranslate-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":53248,"max_output_tokens":4096,"model_type":"translation","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen3-LiveTranslate-Flash is an audio and video translation model based on the Qwen3-Omni architecture. It supports translation between 18 languages, including Chinese, English, Russian, and French. The model can use visual context to improve translation accuracy and outputs both text and speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2999748.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2999751.html) #### **International** If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen3-LiveTranslate-Flash is an audio and video translation model based on the Qwen3-Omni architecture.","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":true},{"id":"qwen3-max-preview","name":"qwen3-max-preview","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3-max","name":"qwen3-max","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":256000,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"],"reasoning_tokens":true},{"id":"qwen3-next-80b-a3b-instruct","name":"qwen3-next-80b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":80,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-next-80b-a3b-thinking","name":"qwen3-next-80b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"model_type":"chat","license":"apache-2.0","parameters":80,"active_parameters":3,"context_window":256000,"reasoning_tokens":true,"page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"qwen3-omni-30b-a3b-captioner","name":"qwen3-omni-30b-a3b-captioner","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":65536,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text","audio"],"output":["text"]},"pricing":{"input":3.81,"output":3.06},"endpoints":["chat"],"license":"apache-2.0","parameters":30,"active_parameters":3,"description":"Qwen3-Omni-Captioner is an open-source model based on Qwen3-Omni. Without any prompts, it automatically generates accurate, comprehensive descriptions for complex audio, ambient sounds, music, film sound effects, and more. It detects speaker emotions, musical elements (such as genre and instruments), sensitive information, and is suitable for audio content analysis, security review, intent recognition, audio editing, and other fields. [Usage](https://www.alibabacloud.com/help/en/document_detail/2980468.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen3-Omni-Captioner is an open-source model built on Qwen3-Omni.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-omni-flash-0915","name":"qwen3-omni-flash-0915","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen3-omni-flash-realtime","name":"qwen3-omni-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":65536,"max_output_tokens":16384,"model_type":"audio","capabilities":{"streaming":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input with built-in VAD (Voice Activity Detection) to automatically detect the start and end of user speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input and includes built-in Voice Activity Detection (VAD) to automatically detect the start and end of user speech.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-omni-flash","name":"qwen3-omni-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","context_window":65536,"max_output_tokens":32768,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Omni accepts multimodal inputs, such as text, images, audio, and video, and generates text or speech responses.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true},{"id":"qwen3-omni-realtime-flash","name":"qwen3-omni-realtime-flash","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwen3-rerank","name":"qwen3-rerank","created_by":"qwen","source":"official","family":"qwen3","model_type":"rerank","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3-tts-flash-realtime","name":"qwen3-tts-flash-realtime","created_by":"qwen","source":"official","family":"qwen3","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen3-tts-flash","name":"qwen3-tts-flash","created_by":"qwen","source":"official","family":"qwen3","model_type":"tts","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"last_updated":"2026-06-03"},{"id":"qwen3-tts-instruct-flash-realtime","name":"qwen3-tts-instruct-flash-realtime","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"tts","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"endpoints":["chat"],"license":"apache-2.0","description":"Supports streaming text input and streaming audio output. It can automatically adjust the speech rate based on the text content and punctuation. [Usage](https://www.alibabacloud.com/help/en/document_detail/2938790.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2938849.html) Qwen3-TTS-Instruct-Flash-Realtime supports [Qwen real-time speech synthesis](https://www.alibabacloud.com/help/en/document_detail/2938790.html#12884a10929p9) and can only use the default voice. It does not support cloned or designed voices. Qwen3-TTS-VD-Realtime supports using voices from [Qwen voice design](https://www.alibabacloud.com/help/en/document_detail/3000986.html) for real-time speech synthesis, but does not support the default voice. Qwen3-TTS-VC-Realtime supports using voices from [Qwen voice cloning](https://www.alibabacloud.com/help/en/document_detail/2975034.html) for real-time speech synthesis, but does not support the default voice. Qwen3-TTS-Flash-Realtime and Qwen-TTS-Realtime can only use the default voice. They do not support cloned or designed voices. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore. ##### **Qwen3-TTS-Instruct-Flash-Realtime**","tagline":"Supports streaming text input and streaming audio output.","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true},{"id":"qwen3-tts-instruct-flash","name":"qwen3-tts-instruct-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3","model_type":"tts","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["audio"]},"endpoints":["chat"],"license":"apache-2.0","description":"Supports mixed-language text input and streaming audio output. [Usage](https://www.alibabacloud.com/help/en/document_detail/2879134.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2881635.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore. ##### **Qwen3-TTS-Instruct-Flash**","tagline":"Supports mixed-language text input and streaming audio output.","page_url":"https://www.alibabacloud.com/help/en/model-studio/tts-model","open_weight":true},{"id":"qwen3-vl-235b-a22b-instruct","name":"qwen3-vl-235b-a22b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":235,"active_parameters":22,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-235b-a22b-thinking","name":"qwen3-vl-235b-a22b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":126976,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.4,"output":4},"endpoints":["chat"],"license":"apache-2.0","parameters":235,"active_parameters":22,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-30b-a3b-instruct","name":"qwen3-vl-30b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":30,"active_parameters":3,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-30b-a3b-thinking","name":"qwen3-vl-30b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":126976,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.2,"output":2.4},"endpoints":["chat"],"license":"apache-2.0","parameters":30,"active_parameters":3,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-32b-instruct","name":"qwen3-vl-32b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-32b-thinking","name":"qwen3-vl-32b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":131072,"max_output_tokens":81920,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.16,"output":0.64},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-8b-instruct","name":"qwen3-vl-8b-instruct","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":129024,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":8,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-8b-thinking","name":"qwen3-vl-8b-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":126976,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"pricing":{"input":0.18,"output":2.1},"endpoints":["chat"],"license":"apache-2.0","parameters":8,"description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-embedding","name":"qwen3-vl-embedding","created_by":"qwen","source":"official","family":"qwen3","model_type":"embed","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":true,"capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3-vl-flash","name":"qwen3-vl-flash","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":81920,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-VL is a text generation model with visual (image) understanding capabilities. It performs OCR (Optical Character Recognition) and further summarizes and infers. For example, it extracts properties from product photos and solves problems based on exercise diagrams. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen-VL models are billed based on the total number of input and output tokens. For rules on calculating image tokens, see [Image and Video Understanding](https://www.alibabacloud.com/help/en/document_detail/2845871.html#7487c7f6eakzl). #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-VL is a text generation model with visual (image) understanding capabilities.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-plus","name":"qwen3-vl-plus","created_by":"qwen","source":"official","last_updated":"2026-04-21","family":"qwen3","context_window":262144,"max_output_tokens":81920,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text","image","video"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-VL is a text generation model with visual (image) understanding capabilities. It performs OCR (Optical Character Recognition) and further summarizes and infers. For example, it extracts properties from product photos and solves problems based on exercise diagrams. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen-VL models are billed based on the total number of input and output tokens. For rules on calculating image tokens, see [Image and Video Understanding](https://www.alibabacloud.com/help/en/document_detail/2845871.html#7487c7f6eakzl). #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-VL is a text generation model with visual (image) understanding capabilities.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwen3-vl-rerank","name":"qwen3-vl-rerank","created_by":"qwen","source":"official","family":"qwen3","model_type":"rerank","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":true,"capabilities":{"streaming":true,"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"qwen3.5-122b-a10b","name":"qwen3.5-122b-a10b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":122,"active_parameters":10,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-27b","name":"qwen3.5-27b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":27,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-35b-a3b","name":"qwen3.5-35b-a3b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":35,"active_parameters":3,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-397b-a17b","name":"qwen3.5-397b-a17b","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":256000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","parameters":397,"active_parameters":17,"description":"Accepts text, image, and video input. Performs on par with Qwen3 Max for plain text tasks---faster and more cost-effective. Offers significant improvements in multimodal capabilities compared to the Qwen3 VL series.","tagline":"Supports text, image, and video inputs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-flash","name":"qwen3.5-flash","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":1000000,"max_output_tokens":64000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.1,"output":0.4},"endpoints":["chat"],"license":"apache-2.0","description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The fastest and most cost-effective model in the Qwen series, ideal for simple tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"tools":["function_calling"]},{"id":"qwen3.5-livetranslate-flash-realtime","name":"qwen3.5-livetranslate-flash-realtime","created_by":"qwen","source":"official","family":"qwen3.5","model_type":"audio","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/s2s-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true},"last_updated":"2026-06-03"},{"id":"qwen3.5-omni-flash-realtime","name":"qwen3.5-omni-flash-realtime","created_by":"qwen","source":"official","family":"qwen3.5","description":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input with built-in VAD (Voice Activity Detection) to automatically detect the start and end of user speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":262144,"max_output_tokens":65536,"model_type":"audio","license":"apache-2.0","tagline":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input and includes built-in VAD (Voice Activity Detection) to automatically detect speech start and end points.[Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client-side events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server-side events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International When the service deployment scope is [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh), data storage is in the **Singapore access region**.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03"},{"id":"qwen3.5-omni-flash","name":"qwen3.5-omni-flash","created_by":"qwen","source":"official","family":"qwen3.5","description":"The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":262144,"model_type":"chat","license":"apache-2.0","tagline":"The Qwen-Omni model accepts multimodal inputs---including text, images, audio, and video---and generates responses in text or speech form.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03"},{"id":"qwen3.5-omni-plus-realtime","name":"qwen3.5-omni-plus-realtime","created_by":"qwen","source":"official","family":"qwen3.5","description":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input with built-in VAD (Voice Activity Detection) to automatically detect the start and end of user speech. [Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":262144,"max_output_tokens":65536,"model_type":"audio","license":"apache-2.0","tagline":"Compared to Qwen-Omni, Qwen-Omni-Realtime supports streaming audio input and includes built-in VAD (Voice Activity Detection) to automatically detect speech start and end points.[Usage](https://www.alibabacloud.com/help/en/document_detail/2880812.html) \\| [Client-side events](https://www.alibabacloud.com/help/en/document_detail/2922854.html) \\| [Server-side events](https://www.alibabacloud.com/help/en/document_detail/2922855.html) #### International When the service deployment scope is [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh), data storage is in the **Singapore access region**.","page_url":"https://www.alibabacloud.com/help/en/model-studio/asr-model","open_weight":true,"capabilities":{"streaming":true,"fine_tuning":true,"tool_call":true,"vision":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","tools":["function_calling"]},{"id":"qwen3.5-omni-plus","name":"qwen3.5-omni-plus","created_by":"qwen","source":"official","family":"qwen3.5","description":"The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. [Usage](https://www.alibabacloud.com/help/en/document_detail/2867839.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":64000,"model_type":"chat","license":"apache-2.0","tagline":"The Qwen-Omni model accepts multimodal inputs---including text, images, audio, and video---and generates responses in text or speech form.","page_url":"https://www.alibabacloud.com/help/en/model-studio/vision-model","open_weight":true,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat"],"last_updated":"2026-06-03"},{"id":"qwen3.5-plus","name":"qwen3.5-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwen3.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"proprietary","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":false,"max_output_tokens":64000,"tools":["function_calling"]},{"id":"qwen3.6-27b","name":"qwen3.6-27b","created_by":"qwen","source":"official","family":"qwen3.6","context_window":229376,"model_type":"chat","license":"apache-2.0","parameters":27,"page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-04-24"},{"id":"qwen3.6-35b-a3b","name":"qwen3.6-35b-a3b","created_by":"qwen","source":"official","family":"qwen3.6","context_window":256000,"model_type":"chat","license":"apache-2.0","parameters":35,"active_parameters":3,"page_url":"https://www.alibabacloud.com/help/en/model-studio/vision-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","max_output_tokens":64000,"tools":["function_calling"]},{"id":"qwen3.6-flash","name":"qwen3.6-flash","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":1000000,"model_type":"chat","license":"apache-2.0","tagline":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","reasoning_tokens":true,"tools":["function_calling"]},{"id":"qwen3.6-max-preview","name":"qwen3.6-max-preview","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":256000,"max_output_tokens":64000,"model_type":"chat","license":"apache-2.0","tagline":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","tools":["function_calling"]},{"id":"qwen3.6-plus","name":"qwen3.6-plus","created_by":"qwen","source":"official","family":"qwen3.6","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","context_window":1000000,"model_type":"chat","license":"apache-2.0","tagline":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-06-03","max_output_tokens":64000,"tools":["function_calling"]},{"id":"qwen3.7-max-preview","name":"qwen3.7-max-preview","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"max_output_tokens":64000,"model_type":"chat","license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3.7-max","name":"qwen3.7-max","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3.7-plus","name":"qwen3.7-plus","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"model_type":"chat","reasoning_tokens":true,"license":"apache-2.0","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-03"},{"id":"qwen3","name":"qwen3","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"model_type":"chat","license":"apache-2.0"},{"id":"qwq-32b-preview","name":"qwq-32b-preview","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwq","context_window":32768,"max_output_tokens":16384,"model_type":"reasoning","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.287,"output":0.861},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"qwq-32b-preview is an experimental research model developed by the Qwen team in 2024, focused on enhancing AI reasoning capabilities, especially in math and programming. See [QwQ official blog](https://qwenlm.github.io/zh/blog/qwq-32b-preview/#%E6%A8%A1%E5%9E%8B%E5%B1%80%E9%99%90%E6%80%A7) for model limitations. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/model-market/detail/qwq-32b-preview) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"qwq-32b-preview is an experimental research model developed by the Qwen team in 2024.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwq-32b","name":"qwq-32b","created_by":"qwen","source":"official","last_updated":"2026-04-03","family":"qwq","context_window":131072,"max_output_tokens":32768,"model_type":"reasoning","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.287,"output":0.861},"endpoints":["chat"],"license":"apache-2.0","parameters":32,"description":"QwQ is a reasoning model trained from Qwen2.5-32B, with significantly enhanced reasoning capabilities through reinforcement learning. Its core metrics (AIME 24/25, LiveCodeBench) and some general metrics (IFEval, LiveBench, etc.) match DeepSeek-R1 full version levels and significantly outperform DeepSeek-R1-Distill-Qwen-32B, which is also based on Qwen2.5-32B. [Usage](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"QwQ is an inference model trained on Qwen2.5-32B.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":true},{"id":"qwq-plus","name":"qwq-plus","created_by":"qwen","source":"official","last_updated":"2026-06-03","family":"qwq","context_window":128000,"max_output_tokens":32768,"model_type":"reasoning","capabilities":{"streaming":true,"reasoning":true,"fine_tuning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"license":"apache-2.0","description":"The QwQ reasoning model, trained on the Qwen2.5 model, uses reinforcement learning to significantly improve its model inference capabilities. The model's core metrics for math and code (AIME 24/25, LiveCodeBench) and general metrics (IFEval, LiveBench) are on par with the full-performance version of DeepSeek-R1. [Usage](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"QwQ is a reasoning model trained on the Qwen2.5 base and significantly enhanced through reinforcement learning.","page_url":"https://www.alibabacloud.com/help/en/model-studio/text-generation-model","open_weight":true,"pricing":{"input":0.8,"output":2.4},"reasoning_tokens":true,"tools":["function_calling"]},{"id":"text-embedding-v3","name":"text-embedding-v3","created_by":"qwen","source":"official","family":"text-embedding","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"text-embedding-v4","name":"text-embedding-v4","created_by":"qwen","source":"official","family":"text-embedding","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"tongyi-embedding-vision-flash","name":"tongyi-embedding-vision-flash","created_by":"qwen","source":"official","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"tongyi-embedding-vision-plus","name":"tongyi-embedding-vision-plus","created_by":"qwen","source":"official","model_type":"embed","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/embedding-rerank-model","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03"},{"id":"wan-pro","name":"wan-pro","created_by":"qwen","source":"official","last_updated":"2026-04-21","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan","model_type":"video","license":"proprietary","description":"Offers standard and professional service modes. Uses a portrait image and reference video to transfer the video subject's actions and expressions to the portrait image, generating a dynamic action video. [API reference](https://www.alibabacloud.com/help/en/document_detail/2981852.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"This service offers Standard and Professional modes.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.1-kf2v-plus","name":"wan2.1-kf2v-plus","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.1","license":"proprietary","description":"The Wan first-and-last-frame image-to-video model generates smooth, fluid videos using just two images---the first and last frames---plus your prompt. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2880649.html) \\| [Try it now](https://wan.video/wanxiang/videoCreation) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"Wan first-and-last-frame video generation models generate smooth, fluid videos from just two input frames, the first and last, plus a prompt.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.1-t2i-plus","name":"wan2.1-t2i-plus","created_by":"qwen","source":"official","family":"wan2.1","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.1-t2i-turbo","name":"wan2.1-t2i-turbo","created_by":"qwen","source":"official","family":"wan2.1","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.1-vace-plus","name":"wan2.1-vace-plus","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.1","license":"proprietary","description":"The Wan video editing model supports multimodal inputs, including text, images, and video, enabling video generation and general editing tasks. [API reference (Video Editing 2.7)](https://www.alibabacloud.com/help/en/document_detail/3021842.html) \\| [API reference (Video Editing 2.1)](https://www.alibabacloud.com/help/en/document_detail/2922183.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"Wan general video editing models support multimodal inputs, including text, images, and video, and perform both video generation and general editing tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-animate-mix","name":"wan2.2-animate-mix","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Offers standard and professional service modes. Uses a portrait image and reference video to replace the main subject in the video with the portrait image, while preserving the original video's scene, lighting, and hue. [API reference](https://www.alibabacloud.com/help/en/document_detail/2982219.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"This service offers Standard and Professional modes.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-animate-move","name":"wan2.2-animate-move","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Offers standard and professional service modes. Uses a portrait image and reference video to transfer the video subject's actions and expressions to the portrait image, generating a dynamic action video. [API reference](https://www.alibabacloud.com/help/en/document_detail/2981852.html) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"This service offers Standard and Professional modes.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-kf2v-flash","name":"wan2.2-kf2v-flash","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"The Wan first-and-last-frame image-to-video model generates smooth, fluid videos using just two images---the first and last frames---plus your prompt. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2880649.html) \\| [Try it now](https://wan.video/wanxiang/videoCreation) ## International If you select the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are restricted to the Chinese mainland. Static data is stored in your selected region. Supported region: China (Beijing).","tagline":"Wan first-and-last-frame video generation models generate smooth, fluid videos from just two input frames, the first and last, plus a prompt.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-s2v-detect","name":"wan2.2-s2v-detect","created_by":"qwen","source":"official","last_updated":"2026-04-03","model_type":"video","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Generate natural talking, singing, or performing videos from a single portrait image and audio. Call the following models in order. [wan2.2-s2v image detection](https://www.alibabacloud.com/help/en/document_detail/2978214.html) \\| [wan2.2-s2v video generation](https://www.alibabacloud.com/help/en/document_detail/2978215.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"You can generate natural-speaking, singing, or performing videos from a single portrait image and audio file.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-s2v","name":"wan2.2-s2v","created_by":"qwen","source":"official","last_updated":"2026-04-03","model_type":"video","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"endpoints":["chat"],"family":"wan2.2","license":"proprietary","description":"Generate natural talking, singing, or performing videos from a single portrait image and audio. Call the following models in order. [wan2.2-s2v image detection](https://www.alibabacloud.com/help/en/document_detail/2978214.html) \\| [wan2.2-s2v video generation](https://www.alibabacloud.com/help/en/document_detail/2978215.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"You can generate natural-speaking, singing, or performing videos from a single portrait image and audio file.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.2-t2i-flash","name":"wan2.2-t2i-flash","created_by":"qwen","source":"official","family":"wan2.2","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.2-t2i-plus","name":"wan2.2-t2i-plus","created_by":"qwen","source":"official","family":"wan2.2","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.5-i2i-preview","name":"wan2.5-i2i-preview","created_by":"qwen","source":"official","last_updated":"2026-06-03","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["video"]},"endpoints":["chat"],"family":"wan2.5","model_type":"video","license":"proprietary","description":"The Wan general image editing 2.5 model supports inputting text, a single image, or multiple images to perform subject-consistent image editing and multi-image fusion creation. [API reference](https://www.alibabacloud.com/help/en/document_detail/2982258.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Wan general image editing 2.5 supports subject-consistent editing using text, a single image, or multiple images.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"wan2.5-t2i-preview","name":"wan2.5-t2i-preview","created_by":"qwen","source":"official","family":"wan2.5","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.6-i2v","name":"wan2.6-i2v","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan image-to-video model uses your input image as the first frame, then generates a video based on your prompt. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2867393.html) \\| [Try it now](https://modelstudio.console.alibabacloud.com/?tab=dashboard#/efm/model_experience_center/vision?currentTab=videoGenerate) ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan image-to-video models use your input image as the first frame, then generate a video based on your prompt, delivering rich artistic styles and cinematic quality.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.6-image","name":"wan2.6-image","created_by":"qwen","source":"official","last_updated":"2026-06-03","model_type":"image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan image generation model supports text-to-image, text-to-series, image-to-series, image editing, and multi-image reference generation, meeting diverse generation and integration needs. [API reference - 2.7](https://www.alibabacloud.com/help/en/document_detail/3026980.html) \\| [API reference - 2.6](https://www.alibabacloud.com/help/en/document_detail/3001143.html) Billing rule: You are charged for the number of **successfully generated images**. Failed attempts are not billed and do not consume your free quota. ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan image generation models support image editing and mixed text-image outputs to meet diverse generation and integration needs.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"wan2.6-r2v","name":"wan2.6-r2v","created_by":"qwen","source":"official","last_updated":"2026-04-22","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text","audio","video"],"output":["text"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan reference-to-video model lets you generate performance videos using characters and voices from reference videos or images. [API reference](https://www.alibabacloud.com/help/en/document_detail/3001146.html) Billing rule: Both input and output videos are billed by **video duration in seconds**. Failed requests are not billed and do not consume your free quota. * Input video duration is capped at **5 seconds** . See [Wan reference-to-video](https://www.alibabacloud.com/help/en/document_detail/3001146.html#f79461ca408qn) for details. * Output video duration equals the **duration of successfully generated video**. ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"The Wan reference-to-video model generates performance videos by referencing a character's appearance from an input video or image, the voice from the input video, and text prompts.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.6-t2i","name":"wan2.6-t2i","created_by":"qwen","source":"official","last_updated":"2026-06-03","model_type":"image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan text-to-image model generates high-quality images from text. [API reference](https://www.alibabacloud.com/help/en/document_detail/2862677.html) \\| [Try online](https://wan.video/wanxiang/creation) ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan text-to-image models generate high-quality images from simple text prompts.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"wan2.6-t2v","name":"wan2.6-t2v","created_by":"qwen","source":"official","last_updated":"2026-04-21","model_type":"video","capabilities":{"streaming":true},"modalities":{"input":["text","audio"],"output":["text"]},"endpoints":["chat"],"family":"wan2.6","license":"proprietary","description":"The Wan text-to-video model generates videos from a single sentence. Videos feature rich artistic styles and cinematic-quality visuals. [API reference](https://www.alibabacloud.com/help/en/document_detail/2865250.html) \\| [Try it now](https://modelstudio.console.alibabacloud.com/?tab=dashboard#/efm/model_experience_center/vision?currentTab=videoGenerate) ## Global If you select the [Global](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide. Static data is stored in your selected region. Supported regions: US (Virginia) and Germany (Frankfurt).","tagline":"Wan text-to-video models generate videos from a single sentence, delivering rich artistic styles and cinematic quality.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false},{"id":"wan2.7-i2v","name":"wan2.7-i2v","created_by":"qwen","source":"official","family":"wan2.7","description":"The Wan image-to-video model has been upgraded to support multimodal inputs (text, image, audio, and video), and can perform three types of tasks: **first-frame-to-video, first-and-last-frame-to-video, and video continuation** . [API reference](https://www.alibabacloud.com/help/en/document_detail/3025059.html) ## International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","model_type":"video","license":"proprietary","tagline":"The Wan image-to-video model has been upgraded to support multimodal inputs (text, image, audio, and video), and can perform three types of tasks: **first-frame-to-video, first-and-last-frame-to-video, and video continuation** .","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text","video"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-04-21"},{"id":"wan2.7-image-pro","name":"wan2.7-image-pro","created_by":"qwen","source":"official","family":"wan2.7","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.7-image","name":"wan2.7-image","created_by":"qwen","source":"official","family":"wan2.7","model_type":"image","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["image"]},"last_updated":"2026-06-03"},{"id":"wan2.7-videoedit","name":"wan2.7-videoedit","created_by":"qwen","source":"official","family":"wan2.7","description":"The Wan video editing model supports multimodal inputs, including text, images, and video, enabling video generation and general editing tasks. [API reference (Video Editing 2.7)](https://www.alibabacloud.com/help/en/document_detail/3021842.html) \\| [API reference (Video Editing 2.1)](https://www.alibabacloud.com/help/en/document_detail/2922183.html) ## International When the service deployment scope is[Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh), data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","model_type":"video","license":"proprietary","tagline":"The Wan video editing model supports multimodal inputs, including text, images, and video, enabling video generation and general editing tasks.","page_url":"https://www.alibabacloud.com/help/en/model-studio/models","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text","video"],"output":["text"]},"endpoints":["chat"],"last_updated":"2026-04-03"},{"id":"wanx2.1-imageedit","name":"wanx2.1-imageedit","created_by":"qwen","source":"official","last_updated":"2026-06-03","model_type":"image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["chat"],"family":"wan","license":"proprietary","description":"The Wan general image editing model supports diverse image editing tasks using simple instructions. Use it for image outpainting, watermark removal, style transfer, image inpainting, and image enhancement. [Usage](https://www.alibabacloud.com/help/en/document_detail/2874281.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2868981.html) **Note** Only the [Chinese mainland](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) service deployment scope is supported. Data storage is in the **Beijing access region**. Model inference compute resources are limited to the Chinese mainland.","tagline":"Wan general image editing models enable diverse image editing tasks using simple instructions.","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false},{"id":"z-image-turbo","name":"z-image-turbo","created_by":"qwen","source":"official","max_output_tokens":1,"model_type":"chat","license":"proprietary","page_url":"https://www.alibabacloud.com/help/en/model-studio/image-model","open_weight":false,"capabilities":{"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"last_updated":"2026-06-03"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/replicate.ts b/packages/npm/src/providers/replicate.ts
index 06cd4f198..132b21726 100644
--- a/packages/npm/src/providers/replicate.ts
+++ b/packages/npm/src/providers/replicate.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"replicate","name":"Replicate","region":"US","headquarters":"San Francisco, CA","founded":2019,"url":"https://replicate.com","api_url":"https://api.replicate.com/v1","docs_url":"https://replicate.com/docs","pricing_url":"https://replicate.com/pricing","description":"Cloud platform for running and publishing machine learning models through API endpoints.","type":"cloud","models_url":"https://replicate.com/explore","openai_compatible":false,"free_tier":false,"github_url":"https://github.com/replicate","twitter_url":"https://x.com/replicate","icon":"","models":[{"id":"55zn56bk8t-art/vishnu","name":"Vishnu","created_by":"55zn56bk8t-art","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/55zn56bk8t-art/vishnu","last_updated":"2026-06-18"},{"id":"851-labs/background-remover","name":"Background Remover","created_by":"851-labs","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/851-labs/background-remover","last_updated":"2026-05-11"},{"id":"adirik/flux-cinestill","name":"Flux Cinestill","created_by":"adirik","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/adirik/flux-cinestill","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"aisha-ai-official/animagine-xl-v4-opt","name":"Animagine Xl V4 Opt","created_by":"aisha-ai-official","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/aisha-ai-official/animagine-xl-v4-opt","last_updated":"2026-06-20"},{"id":"aj718/perkins-alexis","name":"Perkins Alexis","created_by":"aj718","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/aj718/perkins-alexis","last_updated":"2026-05-11"},{"id":"alepfa/glas","name":"Glas","created_by":"alepfa","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alepfa/glas","last_updated":"2026-05-11"},{"id":"alepfa/kollektion","name":"Kollektion","created_by":"alepfa","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alepfa/kollektion","last_updated":"2026-05-11"},{"id":"alepfa/sign","name":"Sign","created_by":"alepfa","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alepfa/sign","last_updated":"2026-05-11"},{"id":"alibaba/happyhorse-1.0","name":"Happyhorse 1.0","created_by":"alibaba","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alibaba/happyhorse-1.0","last_updated":"2026-05-11"},{"id":"andreasjansson/clip-features","name":"Clip Features","created_by":"andreasjansson","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/andreasjansson/clip-features","last_updated":"2026-05-11"},{"id":"anthropic/claude-4.5-sonnet","name":"Claude 4.5 Sonnet","created_by":"anthropic","source":"official","family":"claude-sonnet","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-4.5-sonnet","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"anthropic/claude-fable-5","name":"Claude Fable 5","created_by":"anthropic","source":"official","family":"claude","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-fable-5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-10","description":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","tagline":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","context_window":1000000,"max_output_tokens":128000,"license":"proprietary","open_weight":false},{"id":"anthropic/claude-opus-4.6","name":"Claude Opus 4.6","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-opus-4.6","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"anthropic/claude-opus-4.7","name":"Claude Opus 4.7","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-opus-4.7","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"anthropic/claude-sonnet-4.6","name":"Claude Sonnet 4.6","created_by":"anthropic","source":"official","family":"claude-sonnet","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-sonnet-4.6","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"appmeloncreator/platmoji-beta","name":"Platmoji Beta","created_by":"appmeloncreator","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/appmeloncreator/platmoji-beta","last_updated":"2026-05-11"},{"id":"arielreplicate/robust_video_matting","name":"Robust Video Matting","created_by":"arielreplicate","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/arielreplicate/robust_video_matting","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"arsocekaj/anime-sdxl-v17","name":"Anime Sdxl V17","created_by":"arsocekaj","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/arsocekaj/anime-sdxl-v17","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-03"},{"id":"beautyyuyanli/multilingual-e5-large","name":"Multilingual E5 Large","created_by":"beautyyuyanli","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/beautyyuyanli/multilingual-e5-large","last_updated":"2026-05-11"},{"id":"bencevans/megadetector-v1000-redwood","name":"Megadetector V1000 Redwood","created_by":"bencevans","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bencevans/megadetector-v1000-redwood","last_updated":"2026-06-10"},{"id":"birgit-ploeger/birgit-portrait","name":"Birgit Portrait","created_by":"birgit-ploeger","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/birgit-ploeger/birgit-portrait","last_updated":"2026-06-03"},{"id":"black-forest-labs/flux-2-flex","name":"Flux 2 Flex","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-2-flex","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"FLUX.2 variant with adjustable steps and guidance, tuned for typography and editing workflows. Up to 8 references via API.","tagline":"Flexible FLUX.2 for typography and editing.","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-2-klein-4b","name":"Flux 2 Klein 4b","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","parameters":4,"page_url":"https://replicate.com/black-forest-labs/flux-2-klein-4b","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Lightweight 4B-parameter FLUX.2 flow model for sub-second inference; ~13 GB VRAM. Open base weights under Apache 2.0.","tagline":"Fastest, most lightweight FLUX.2.","license":"apache-2.0","open_weight":true},{"id":"black-forest-labs/flux-2-max","name":"Flux 2 Max","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-2-max","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Highest-quality FLUX.2 model with web grounding search, up to 8 references via API and 10 in playground.","tagline":"Highest quality FLUX.2.","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-2-pro","name":"Flux 2 Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-2-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Production-grade default model for FLUX.2 text-to-image generation and image editing. Supports up to 10 reference images and 4MP output.","tagline":"Recommended default for image generation and editing.","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-canny-pro","name":"Flux Canny Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-canny-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"black-forest-labs/flux-depth-pro","name":"Flux Depth Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-depth-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"black-forest-labs/flux-dev","name":"Flux Dev","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-dev","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"12B-parameter FLUX.1 [dev] model for text-to-image generation. Open weights for non-commercial use; also exposed via BFL API for hosted inference.","tagline":"FLUX.1 12B open-weight dev model.","license":"flux-1-dev-non-commercial","open_weight":true},{"id":"black-forest-labs/flux-kontext-max","name":"Flux Kontext Max","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-kontext-max","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Higher-quality variant of Kontext with stronger editing fidelity. Legacy; superseded by FLUX.2 [max] for new projects.","tagline":"Legacy FLUX.1 Kontext [max].","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-kontext-pro","name":"Flux Kontext Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-kontext-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Context-aware image editing and text-to-image. Legacy FLUX.1 line; BFL recommends FLUX.2 [pro] for new projects.","tagline":"Legacy FLUX.1 Kontext [pro].","license":"proprietary","open_weight":false},{"id":"bytedance/dolphin","name":"Dolphin","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/dolphin","last_updated":"2026-05-11"},{"id":"bytedance/hyper-flux-8step","name":"Hyper Flux 8step","created_by":"bytedance","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/bytedance/hyper-flux-8step","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"bytedance/seedance-2.0","name":"Seedance 2.0","created_by":"bytedance","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/bytedance/seedance-2.0","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11","description":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","tagline":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","capabilities":{"vision":true}},{"id":"bytedance/seedream-4.5","name":"Seedream 4.5","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/seedream-4.5","last_updated":"2026-05-11","description":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","tagline":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seedream-4","name":"Seedream 4","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/seedream-4","last_updated":"2026-05-11"},{"id":"bytedance/seedream-5-lite","name":"Seedream 5 Lite","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/seedream-5-lite","last_updated":"2026-05-11"},{"id":"ccchot-osk103/buacat","name":"Buacat","created_by":"ccchot-osk103","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ccchot-osk103/buacat","last_updated":"2026-05-11"},{"id":"ccchot-osk103/moneycat","name":"Moneycat","created_by":"ccchot-osk103","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ccchot-osk103/moneycat","last_updated":"2026-05-11"},{"id":"ccchot-osk103/pai_qwen_21102568","name":"Pai Qwen 21102568","created_by":"ccchot-osk103","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ccchot-osk103/pai_qwen_21102568","last_updated":"2026-05-11"},{"id":"charlesmccarthy/addwatermark","name":"Addwatermark","created_by":"charlesmccarthy","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/charlesmccarthy/addwatermark","last_updated":"2026-05-11"},{"id":"cjwbw/anything-v4.0","name":"Anything V4.0","created_by":"cjwbw","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/cjwbw/anything-v4.0","last_updated":"2026-05-11"},{"id":"cjwbw/semantic-segment-anything","name":"Semantic Segment Anything","created_by":"cjwbw","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/cjwbw/semantic-segment-anything","last_updated":"2026-05-11"},{"id":"codeplugtech/face-swap","name":"Face Swap","created_by":"codeplugtech","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/codeplugtech/face-swap","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"colinhughes2121/ai-anime-villain","name":"Ai Anime Villain","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-anime-villain","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-charcoal-portrait","name":"Ai Charcoal Portrait","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-charcoal-portrait","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-pop-art-stylizer","name":"Ai Pop Art Stylizer","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-pop-art-stylizer","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-tarot-card-stylizer","name":"Ai Tarot Card Stylizer","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-tarot-card-stylizer","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-vintage-film-photo","name":"Ai Vintage Film Photo","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-vintage-film-photo","last_updated":"2026-06-03"},{"id":"daanelson/imagebind","name":"Imagebind","created_by":"daanelson","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/daanelson/imagebind","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"damdam775/portraits_dialogues","name":"Portraits Dialogues","created_by":"damdam775","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/damdam775/portraits_dialogues","last_updated":"2026-05-11"},{"id":"dannyboy2323/character-daddanny23","name":"Character Daddanny23","created_by":"dannyboy2323","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/dannyboy2323/character-daddanny23","last_updated":"2026-06-10"},{"id":"dannyboy2323/character-danielbull23","name":"Character Danielbull23","created_by":"dannyboy2323","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/dannyboy2323/character-danielbull23","last_updated":"2026-06-10"},{"id":"dannyboy2323/character-hater","name":"Character Hater","created_by":"dannyboy2323","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/dannyboy2323/character-hater","last_updated":"2026-06-10"},{"id":"datacte/proteus-v0.2","name":"Proteus V0.2","created_by":"datacte","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/datacte/proteus-v0.2","last_updated":"2026-05-11"},{"id":"datacte/proteus-v0.3","name":"Proteus V0.3","created_by":"datacte","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/datacte/proteus-v0.3","last_updated":"2026-05-11"},{"id":"datalab-to/marker","name":"Marker","created_by":"datalab-to","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/datalab-to/marker","last_updated":"2026-05-11"},{"id":"elevenlabs/music","name":"Music","created_by":"elevenlabs","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/elevenlabs/music","last_updated":"2026-05-11"},{"id":"elevenlabs/scribe-v2","name":"Scribe V2","created_by":"elevenlabs","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/elevenlabs/scribe-v2","last_updated":"2026-05-11"},{"id":"falcons-ai/nsfw_image_detection","name":"Nsfw Image Detection","created_by":"falcons-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/falcons-ai/nsfw_image_detection","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fictions-ai/autocaption","name":"Autocaption","created_by":"fictions-ai","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fictions-ai/autocaption","last_updated":"2026-05-11"},{"id":"fishwowater/trellis2","name":"Trellis2","created_by":"fishwowater","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fishwowater/trellis2","last_updated":"2026-05-11"},{"id":"flux-kontext-apps/restore-image","name":"Restore Image","created_by":"flux-kontext-apps","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/flux-kontext-apps/restore-image","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fofr/expression-editor","name":"Expression Editor","created_by":"fofr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fofr/expression-editor","last_updated":"2026-06-10"},{"id":"fofr/face-swap-with-ideogram","name":"Face Swap With Ideogram","created_by":"fofr","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/fofr/face-swap-with-ideogram","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fofr/flux-mona-lisa","name":"Flux Mona Lisa","created_by":"fofr","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/fofr/flux-mona-lisa","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fofr/latent-consistency-model","name":"Latent Consistency Model","created_by":"fofr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fofr/latent-consistency-model","last_updated":"2026-05-11"},{"id":"fofr/prompt-classifier","name":"Prompt Classifier","created_by":"fofr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fofr/prompt-classifier","last_updated":"2026-06-03"},{"id":"fofr/sdxl-emoji","name":"Sdxl Emoji","created_by":"fofr","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/fofr/sdxl-emoji","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"furkkurt/vector-blog-thumbnails","name":"Vector Blog Thumbnails","created_by":"furkkurt","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/furkkurt/vector-blog-thumbnails","last_updated":"2026-05-11"},{"id":"google/gemini-3-flash","name":"Gemini 3 Flash","created_by":"google","source":"official","family":"gemini-3","status":"active","model_type":"chat","page_url":"https://replicate.com/google/gemini-3-flash","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemini-3.1-flash-tts","name":"Gemini 3.1 Flash Tts","created_by":"google","source":"official","family":"gemini-3.1","status":"active","model_type":"audio","page_url":"https://replicate.com/google/gemini-3.1-flash-tts","last_updated":"2026-05-11"},{"id":"google/gemini-3.1-pro","name":"Gemini 3.1 Pro","created_by":"google","source":"official","family":"gemini-3.1","status":"active","model_type":"chat","page_url":"https://replicate.com/google/gemini-3.1-pro","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemini-3.5-flash","name":"Gemini 3.5 Flash","created_by":"google","source":"official","family":"gemini-3.5","status":"active","model_type":"chat","page_url":"https://replicate.com/google/gemini-3.5-flash","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-20","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/imagen-4","name":"Imagen 4","created_by":"google","source":"official","family":"imagen-4","status":"active","model_type":"image","page_url":"https://replicate.com/google/imagen-4","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/nano-banana-2","name":"Nano Banana 2","created_by":"google","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/google/nano-banana-2","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/nano-banana","name":"Nano Banana","created_by":"google","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/google/nano-banana","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/upscaler","name":"Upscaler","created_by":"google","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/google/upscaler","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/veo-3.1-lite","name":"Veo 3.1 Lite","created_by":"google","source":"official","family":"veo-3.1","status":"active","model_type":"video","page_url":"https://replicate.com/google/veo-3.1-lite","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"hyper3d/rodin","name":"Rodin","created_by":"hyper3d","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/hyper3d/rodin","last_updated":"2026-05-11"},{"id":"ideogram-ai/ideogram-character","name":"Ideogram Character","created_by":"ideogram-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/ideogram-ai/ideogram-character","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"ideogram-ai/ideogram-v4-balanced","name":"Ideogram V4 Balanced","created_by":"ideogram-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/ideogram-ai/ideogram-v4-balanced","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-10"},{"id":"ideogram-ai/ideogram-v4-quality","name":"Ideogram V4 Quality","created_by":"ideogram-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/ideogram-ai/ideogram-v4-quality","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-10"},{"id":"inworld/realtime-tts-1.5-max","name":"Realtime Tts 1.5 Max","created_by":"inworld","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/inworld/realtime-tts-1.5-max","last_updated":"2026-05-11"},{"id":"inworld/realtime-tts-1.5-mini","name":"Realtime Tts 1.5 Mini","created_by":"inworld","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/inworld/realtime-tts-1.5-mini","last_updated":"2026-05-11"},{"id":"inworld/realtime-tts-2","name":"Realtime Tts 2","created_by":"inworld","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/inworld/realtime-tts-2","last_updated":"2026-05-11"},{"id":"jaaari/kokoro-82m","name":"Kokoro 82m","created_by":"jaaari","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/jaaari/kokoro-82m","last_updated":"2026-05-11"},{"id":"jilin-dbgr/icetea","name":"Icetea","created_by":"jilin-dbgr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/jilin-dbgr/icetea","last_updated":"2026-06-20"},{"id":"krea/krea-2-large","name":"Krea 2 Large","created_by":"krea","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/krea/krea-2-large","last_updated":"2026-06-03"},{"id":"krea/krea-2-medium","name":"Krea 2 Medium","created_by":"krea","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/krea/krea-2-medium","last_updated":"2026-06-03"},{"id":"krthr/clip-embeddings","name":"Clip Embeddings","created_by":"krthr","source":"official","status":"active","model_type":"embed","page_url":"https://replicate.com/krthr/clip-embeddings","modalities":{"input":["text"],"output":["text"]},"endpoints":["embeddings"],"last_updated":"2026-06-18"},{"id":"kwaivgi/kling-v3-omni-video","name":"Kling V3 Omni Video","created_by":"kwaivgi","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/kwaivgi/kling-v3-omni-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"lucataco/deepseek-ocr","name":"Deepseek Ocr","created_by":"lucataco","source":"official","family":"deepseek","status":"active","model_type":"chat","page_url":"https://replicate.com/lucataco/deepseek-ocr","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"lucataco/gpt-oss-safeguard-20b","name":"Gpt Oss Safeguard 20b","created_by":"lucataco","source":"official","family":"gpt-oss","status":"active","model_type":"chat","parameters":20,"page_url":"https://replicate.com/lucataco/gpt-oss-safeguard-20b","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"lucataco/kontext-realearth","name":"Kontext Realearth","created_by":"lucataco","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/lucataco/kontext-realearth","last_updated":"2026-05-11"},{"id":"lucataco/motif-video","name":"Motif Video","created_by":"lucataco","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/lucataco/motif-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"lucataco/remove-bg","name":"Remove Bg","created_by":"lucataco","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/lucataco/remove-bg","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"luma/modify-video","name":"Modify Video","created_by":"luma","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/luma/modify-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"luma/reframe-video","name":"Reframe Video","created_by":"luma","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/luma/reframe-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"lxcxsito/stargirlz","name":"Stargirlz","created_by":"lxcxsito","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/lxcxsito/stargirlz","last_updated":"2026-06-18"},{"id":"m1guelpf/nsfw-filter","name":"Nsfw Filter","created_by":"m1guelpf","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/m1guelpf/nsfw-filter","last_updated":"2026-05-11"},{"id":"maikocode/ascii-style","name":"Ascii Style","created_by":"maikocode","source":"official","status":"active","model_type":"code","page_url":"https://replicate.com/maikocode/ascii-style","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"megvii-research/nafnet","name":"Nafnet","created_by":"megvii-research","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/megvii-research/nafnet","last_updated":"2026-05-11"},{"id":"meta/llama-guard-4-12b","name":"Llama Guard 4 12b","created_by":"meta","source":"official","family":"llama-guard","status":"active","model_type":"chat","parameters":12,"page_url":"https://replicate.com/meta/llama-guard-4-12b","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","license":"llama4","open_weight":true},{"id":"meta/sam-2-video","name":"Sam 2 Video","created_by":"meta","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/meta/sam-2-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"miike-ai/flux-ico","name":"Flux Ico","created_by":"miike-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/miike-ai/flux-ico","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"minimax/music-1.5","name":"Music 1.5","created_by":"minimax","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/minimax/music-1.5","last_updated":"2026-05-11"},{"id":"minimax/music-2.5","name":"Music 2.5","created_by":"minimax","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/minimax/music-2.5","last_updated":"2026-05-11"},{"id":"minimax/music-2.6","name":"Music 2.6","created_by":"minimax","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/minimax/music-2.6","last_updated":"2026-05-11"},{"id":"mptamilselvan/download-media","name":"Download Media","created_by":"mptamilselvan","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/mptamilselvan/download-media","last_updated":"2026-05-11"},{"id":"mptamilselvan/text-to-voice","name":"Text To Voice","created_by":"mptamilselvan","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/mptamilselvan/text-to-voice","last_updated":"2026-05-11"},{"id":"myaiteam2/website-scrapper","name":"Website Scrapper","created_by":"myaiteam2","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/myaiteam2/website-scrapper","last_updated":"2026-06-03"},{"id":"naimish-gami/face-swapper","name":"Face Swapper","created_by":"naimish-gami","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/naimish-gami/face-swapper","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-18"},{"id":"nicegen-ai/nsfw-filter-for-portrait","name":"Nsfw Filter For Portrait","created_by":"nicegen-ai","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/nicegen-ai/nsfw-filter-for-portrait","last_updated":"2026-06-18"},{"id":"oitoito/depth-pro","name":"Depth Pro","created_by":"oitoito","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/oitoito/depth-pro","last_updated":"2026-06-10"},{"id":"oitoito/mask2former","name":"Mask2former","created_by":"oitoito","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/oitoito/mask2former","last_updated":"2026-06-10"},{"id":"openai/gpt-4o-mini","name":"Gpt 4o Mini","created_by":"openai","source":"official","family":"gpt-4o","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-4o-mini","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-4o mini (“o” for “omni”) is a fast, affordable small model for focused tasks.","tagline":"Fast, affordable small model for focused tasks","context_window":128000,"max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-transcribe","name":"Gpt 4o Transcribe","created_by":"openai","source":"official","family":"gpt-4o","status":"active","model_type":"transcription","page_url":"https://replicate.com/openai/gpt-4o-transcribe","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11","description":"GPT-4o Transcribe is a speech-to-text model that uses GPT-4o to transcribe audio.","tagline":"Speech-to-text model powered by GPT-4o","context_window":16000,"max_output_tokens":2000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-nano","name":"Gpt 5 Nano","created_by":"openai","source":"official","family":"gpt-5","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-5-nano","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for","tagline":"Fastest, most cost-efficient version of GPT-5","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2","name":"Gpt 5.2","created_by":"openai","source":"official","family":"gpt-5.2","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-5.2","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5","name":"Gpt 5","created_by":"openai","source":"official","family":"gpt-5","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-image-1.5","name":"Gpt Image 1.5","created_by":"openai","source":"official","family":"gpt-image-1.5","status":"active","model_type":"image","page_url":"https://replicate.com/openai/gpt-image-1.5","capabilities":{"vision":true,"streaming":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"GPT Image 1.5 is our previous image generation model, with better instruction following and adherence to prompts. Learn more in our [GPT Image 1.5 usage guide](/api/docs/guides/image-generation).","tagline":"Our previous image generation model","license":"proprietary","open_weight":false},{"id":"openai/gpt-image-2","name":"Gpt Image 2","created_by":"openai","source":"official","family":"gpt-image-2","status":"active","model_type":"image","page_url":"https://replicate.com/openai/gpt-image-2","capabilities":{"vision":true,"streaming":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"GPT Image 2 is our state-of-the-art image generation model for fast, high-quality image generation and editing. It supports flexible image sizes and high-fidelity image inputs. Learn more in our [image generation guide](/api/docs/guides/image-generation), or see the [pricing page](/api/docs/pricing#image-generation) and [image generation calculator](/api/docs/guides/image-generation#calculating-costs) for cost estimates.","tagline":"State-of-the-art image generation model","license":"proprietary","open_weight":false},{"id":"philz1337x/clarity-pro-upscaler","name":"Clarity Pro Upscaler","created_by":"philz1337x","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/philz1337x/clarity-pro-upscaler","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"philz1337x/crystal-video-upscaler","name":"Crystal Video Upscaler","created_by":"philz1337x","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/philz1337x/crystal-video-upscaler","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"piddnad/ddcolor","name":"Ddcolor","created_by":"piddnad","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/piddnad/ddcolor","last_updated":"2026-05-11"},{"id":"prunaai/p-image-edit","name":"P Image Edit","created_by":"prunaai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/prunaai/p-image-edit","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-03"},{"id":"prunaai/p-image-upscale","name":"P Image Upscale","created_by":"prunaai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/prunaai/p-image-upscale","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"prunaai/p-video-animate","name":"P Video Animate","created_by":"prunaai","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/prunaai/p-video-animate","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-18"},{"id":"prunaai/p-video-avatar","name":"P Video Avatar","created_by":"prunaai","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/prunaai/p-video-avatar","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"prunaai/p-video","name":"P Video","created_by":"prunaai","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/prunaai/p-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"prunaai/z-image-turbo","name":"Z Image Turbo","created_by":"prunaai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/prunaai/z-image-turbo","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-tts","name":"Qwen3 Tts","created_by":"alibaba","source":"official","family":"qwen3","status":"active","model_type":"audio","page_url":"https://replicate.com/qwen/qwen3-tts","last_updated":"2026-05-11"},{"id":"rafaelgalle/whisper-diarization-advanced","name":"Whisper Diarization Advanced","created_by":"rafaelgalle","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/rafaelgalle/whisper-diarization-advanced","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"replicate/train-rvc-model","name":"Train Rvc Model","created_by":"replicate","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/replicate/train-rvc-model","last_updated":"2026-05-11"},{"id":"runwayml/aleph-2","name":"Aleph 2","created_by":"runwayml","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/runwayml/aleph-2","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-10"},{"id":"runwayml/gen-4.5","name":"Gen 4.5","created_by":"runwayml","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/runwayml/gen-4.5","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"runwayml/gen4-image","name":"Gen4 Image","created_by":"runwayml","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/runwayml/gen4-image","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"schananas/grounded_sam","name":"Grounded Sam","created_by":"schananas","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/schananas/grounded_sam","last_updated":"2026-05-11"},{"id":"shreejalmaharjan-27/tiktok-short-captions","name":"Tiktok Short Captions","created_by":"shreejalmaharjan-27","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/shreejalmaharjan-27/tiktok-short-captions","last_updated":"2026-05-11"},{"id":"smokeyang/transition_generator","name":"Transition Generator","created_by":"smokeyang","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/smokeyang/transition_generator","last_updated":"2026-06-18"},{"id":"sourceful/riverflow-v2.5-pro","name":"Riverflow V2.5 Pro","created_by":"sourceful","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sourceful/riverflow-v2.5-pro","last_updated":"2026-06-18"},{"id":"sprited/birefnet-toonout","name":"Birefnet Toonout","created_by":"sprited","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sprited/birefnet-toonout","last_updated":"2026-06-18"},{"id":"sprited/birefnet-video","name":"Birefnet Video","created_by":"sprited","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/sprited/birefnet-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-18"},{"id":"sprited/birefnet","name":"Birefnet","created_by":"sprited","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sprited/birefnet","last_updated":"2026-06-18"},{"id":"spuuntries/dfk3-queeree","name":"Dfk3 Queeree","created_by":"spuuntries","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/spuuntries/dfk3-queeree","last_updated":"2026-05-11"},{"id":"sync/lipsync-2-pro","name":"Lipsync 2 Pro","created_by":"sync","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sync/lipsync-2-pro","last_updated":"2026-05-11"},{"id":"syncodeofficial/noah-biblical-craft","name":"Noah Biblical Craft","created_by":"syncodeofficial","source":"official","status":"active","model_type":"code","page_url":"https://replicate.com/syncodeofficial/noah-biblical-craft","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"tattzy25/heavy_hand_dark","name":"Heavy Hand Dark","created_by":"tattzy25","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/tattzy25/heavy_hand_dark","last_updated":"2026-06-18"},{"id":"tencent/hunyuan-3d-3.1","name":"Hunyuan 3d 3.1","created_by":"tencent","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/tencent/hunyuan-3d-3.1","last_updated":"2026-05-11"},{"id":"thomasmol/whisper-diarization","name":"Whisper Diarization","created_by":"thomasmol","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/thomasmol/whisper-diarization","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"topazlabs/video-upscale","name":"Video Upscale","created_by":"topazlabs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/topazlabs/video-upscale","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"vaibhavs10/incredibly-fast-whisper","name":"Incredibly Fast Whisper","created_by":"vaibhavs10","source":"official","status":"active","model_type":"transcription","page_url":"https://replicate.com/vaibhavs10/incredibly-fast-whisper","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"valllllex/cartoonia_3d","name":"Cartoonia 3d","created_by":"valllllex","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/valllllex/cartoonia_3d","last_updated":"2026-05-11"},{"id":"victor-upmeet/whisperx","name":"Whisperx","created_by":"victor-upmeet","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/victor-upmeet/whisperx","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"vm6eji6m4/whisper-chinese-pro","name":"Whisper Chinese Pro","created_by":"vm6eji6m4","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/vm6eji6m4/whisper-chinese-pro","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-06-03"},{"id":"vufinder/sam3","name":"Sam3","created_by":"vufinder","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/vufinder/sam3","last_updated":"2026-06-18"},{"id":"wan-video/wan-2.2-t2v-fast","name":"Wan 2.2 T2v Fast","created_by":"wan-video","source":"official","family":"wan","status":"active","model_type":"video","page_url":"https://replicate.com/wan-video/wan-2.2-t2v-fast","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"wan-video/wan-2.5-i2v-fast","name":"Wan 2.5 I2v Fast","created_by":"wan-video","source":"official","family":"wan","status":"active","model_type":"video","page_url":"https://replicate.com/wan-video/wan-2.5-i2v-fast","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"wan-video/wan-2.5-t2v-fast","name":"Wan 2.5 T2v Fast","created_by":"wan-video","source":"official","family":"wan","status":"active","model_type":"video","page_url":"https://replicate.com/wan-video/wan-2.5-t2v-fast","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"xai/grok-4","name":"Grok 4","created_by":"xai","source":"official","family":"grok-4","status":"active","model_type":"chat","page_url":"https://replicate.com/xai/grok-4","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"xai/grok-imagine-image-quality","name":"Grok Imagine Image Quality","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"image","page_url":"https://replicate.com/xai/grok-imagine-image-quality","capabilities":{"vision":true,"streaming":true,"batch":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false},{"id":"xai/grok-imagine-video-1.5","name":"Grok Imagine Video 1.5","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"video","page_url":"https://replicate.com/xai/grok-imagine-video-1.5","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"batch":true,"vision":true}},{"id":"xai/grok-imagine-video-extension","name":"Grok Imagine Video Extension","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"video","page_url":"https://replicate.com/xai/grok-imagine-video-extension","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"xai/grok-imagine-video","name":"Grok Imagine Video","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"video","page_url":"https://replicate.com/xai/grok-imagine-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true}},{"id":"yangzhou0011/layoutgrok-qwen3-vl-cp500","name":"Layoutgrok Qwen3 Vl Cp500","created_by":"yangzhou0011","source":"official","status":"active","model_type":"chat","page_url":"https://replicate.com/yangzhou0011/layoutgrok-qwen3-vl-cp500","capabilities":{"vision":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-18"},{"id":"zsxkib/create-rvc-dataset","name":"Create Rvc Dataset","created_by":"zsxkib","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/zsxkib/create-rvc-dataset","last_updated":"2026-05-11"},{"id":"zsxkib/multitalk","name":"Multitalk","created_by":"zsxkib","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/zsxkib/multitalk","last_updated":"2026-05-11"},{"id":"zsxkib/realistic-voice-cloning","name":"Realistic Voice Cloning","created_by":"zsxkib","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/zsxkib/realistic-voice-cloning","last_updated":"2026-05-11"}]};
+export const provider: ProviderWithModels = {"id":"replicate","name":"Replicate","region":"US","headquarters":"San Francisco, CA","founded":2019,"url":"https://replicate.com","api_url":"https://api.replicate.com/v1","docs_url":"https://replicate.com/docs","pricing_url":"https://replicate.com/pricing","description":"Cloud platform for running and publishing machine learning models through API endpoints.","type":"cloud","models_url":"https://replicate.com/explore","openai_compatible":false,"free_tier":false,"github_url":"https://github.com/replicate","twitter_url":"https://x.com/replicate","icon":"","models":[{"id":"0xdino/cyberrealistic-pony-v18","name":"Cyberrealistic Pony V18","created_by":"0xdino","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/0xdino/cyberrealistic-pony-v18","last_updated":"2026-06-29"},{"id":"55zn56bk8t-art/vishnu","name":"Vishnu","created_by":"55zn56bk8t-art","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/55zn56bk8t-art/vishnu","last_updated":"2026-06-18"},{"id":"851-labs/background-remover","name":"Background Remover","created_by":"851-labs","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/851-labs/background-remover","last_updated":"2026-05-11"},{"id":"adirik/flux-cinestill","name":"Flux Cinestill","created_by":"adirik","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/adirik/flux-cinestill","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"aisha-ai-official/animagine-xl-v4-opt","name":"Animagine Xl V4 Opt","created_by":"aisha-ai-official","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/aisha-ai-official/animagine-xl-v4-opt","last_updated":"2026-06-20"},{"id":"aj718/perkins-alexis","name":"Perkins Alexis","created_by":"aj718","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/aj718/perkins-alexis","last_updated":"2026-05-11"},{"id":"alepfa/glas","name":"Glas","created_by":"alepfa","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alepfa/glas","last_updated":"2026-05-11"},{"id":"alepfa/kollektion","name":"Kollektion","created_by":"alepfa","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alepfa/kollektion","last_updated":"2026-05-11"},{"id":"alepfa/sign","name":"Sign","created_by":"alepfa","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alepfa/sign","last_updated":"2026-05-11"},{"id":"alibaba/happyhorse-1.0","name":"Happyhorse 1.0","created_by":"alibaba","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alibaba/happyhorse-1.0","last_updated":"2026-05-11"},{"id":"alibaba/happyhorse-1.1","name":"Happyhorse 1.1","created_by":"alibaba","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/alibaba/happyhorse-1.1","last_updated":"2026-06-29"},{"id":"andreasjansson/clip-features","name":"Clip Features","created_by":"andreasjansson","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/andreasjansson/clip-features","last_updated":"2026-05-11"},{"id":"anthropic/claude-4.5-sonnet","name":"Claude 4.5 Sonnet","created_by":"anthropic","source":"official","family":"claude-sonnet","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-4.5-sonnet","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"anthropic/claude-fable-5","name":"Claude Fable 5","created_by":"anthropic","source":"official","family":"claude","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-fable-5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true,"reasoning":true,"structured_output":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-10","description":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","tagline":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","context_window":1000000,"max_output_tokens":128000,"license":"proprietary","open_weight":false},{"id":"anthropic/claude-opus-4.6","name":"Claude Opus 4.6","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-opus-4.6","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"anthropic/claude-opus-4.7","name":"Claude Opus 4.7","created_by":"anthropic","source":"official","family":"claude-opus","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-opus-4.7","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"anthropic/claude-sonnet-4.6","name":"Claude Sonnet 4.6","created_by":"anthropic","source":"official","family":"claude-sonnet","status":"active","model_type":"chat","page_url":"https://replicate.com/anthropic/claude-sonnet-4.6","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"appmeloncreator/platmoji-beta","name":"Platmoji Beta","created_by":"appmeloncreator","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/appmeloncreator/platmoji-beta","last_updated":"2026-05-11"},{"id":"arielreplicate/robust_video_matting","name":"Robust Video Matting","created_by":"arielreplicate","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/arielreplicate/robust_video_matting","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"arsocekaj/anime-sdxl-v17","name":"Anime Sdxl V17","created_by":"arsocekaj","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/arsocekaj/anime-sdxl-v17","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-03"},{"id":"beautyyuyanli/multilingual-e5-large","name":"Multilingual E5 Large","created_by":"beautyyuyanli","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/beautyyuyanli/multilingual-e5-large","last_updated":"2026-05-11"},{"id":"bencevans/megadetector-v1000-redwood","name":"Megadetector V1000 Redwood","created_by":"bencevans","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bencevans/megadetector-v1000-redwood","last_updated":"2026-06-10"},{"id":"birgit-ploeger/birgit-portrait","name":"Birgit Portrait","created_by":"birgit-ploeger","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/birgit-ploeger/birgit-portrait","last_updated":"2026-06-03"},{"id":"black-forest-labs/flux-2-flex","name":"Flux 2 Flex","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-2-flex","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"FLUX.2 variant with adjustable steps and guidance, tuned for typography and editing workflows. Up to 8 references via API.","tagline":"Flexible FLUX.2 for typography and editing.","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-2-klein-4b","name":"Flux 2 Klein 4b","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","parameters":4,"page_url":"https://replicate.com/black-forest-labs/flux-2-klein-4b","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Lightweight 4B-parameter FLUX.2 flow model for sub-second inference; ~13 GB VRAM. Open base weights under Apache 2.0.","tagline":"Fastest, most lightweight FLUX.2.","license":"apache-2.0","open_weight":true},{"id":"black-forest-labs/flux-2-max","name":"Flux 2 Max","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-2-max","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Highest-quality FLUX.2 model with web grounding search, up to 8 references via API and 10 in playground.","tagline":"Highest quality FLUX.2.","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-2-pro","name":"Flux 2 Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-2-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Production-grade default model for FLUX.2 text-to-image generation and image editing. Supports up to 10 reference images and 4MP output.","tagline":"Recommended default for image generation and editing.","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-canny-pro","name":"Flux Canny Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-canny-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"black-forest-labs/flux-depth-pro","name":"Flux Depth Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-depth-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"black-forest-labs/flux-dev","name":"Flux Dev","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-dev","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"12B-parameter FLUX.1 [dev] model for text-to-image generation. Open weights for non-commercial use; also exposed via BFL API for hosted inference.","tagline":"FLUX.1 12B open-weight dev model.","license":"flux-1-dev-non-commercial","open_weight":true},{"id":"black-forest-labs/flux-kontext-max","name":"Flux Kontext Max","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-kontext-max","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Higher-quality variant of Kontext with stronger editing fidelity. Legacy; superseded by FLUX.2 [max] for new projects.","tagline":"Legacy FLUX.1 Kontext [max].","license":"proprietary","open_weight":false},{"id":"black-forest-labs/flux-kontext-pro","name":"Flux Kontext Pro","created_by":"black-forest-labs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/black-forest-labs/flux-kontext-pro","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"Context-aware image editing and text-to-image. Legacy FLUX.1 line; BFL recommends FLUX.2 [pro] for new projects.","tagline":"Legacy FLUX.1 Kontext [pro].","license":"proprietary","open_weight":false},{"id":"bytedance/dolphin","name":"Dolphin","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/dolphin","last_updated":"2026-05-11"},{"id":"bytedance/hyper-flux-8step","name":"Hyper Flux 8step","created_by":"bytedance","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/bytedance/hyper-flux-8step","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"bytedance/seedance-2.0-mini","name":"Seedance 2.0 Mini","created_by":"bytedance","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/bytedance/seedance-2.0-mini","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-29"},{"id":"bytedance/seedance-2.0","name":"Seedance 2.0","created_by":"bytedance","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/bytedance/seedance-2.0","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11","description":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","tagline":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","capabilities":{"vision":true}},{"id":"bytedance/seedream-4.5","name":"Seedream 4.5","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/seedream-4.5","last_updated":"2026-05-11","description":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","tagline":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seedream-4","name":"Seedream 4","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/seedream-4","last_updated":"2026-05-11"},{"id":"bytedance/seedream-5-lite","name":"Seedream 5 Lite","created_by":"bytedance","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/bytedance/seedream-5-lite","last_updated":"2026-05-11"},{"id":"ccchot-osk103/buacat","name":"Buacat","created_by":"ccchot-osk103","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ccchot-osk103/buacat","last_updated":"2026-05-11"},{"id":"ccchot-osk103/moneycat","name":"Moneycat","created_by":"ccchot-osk103","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ccchot-osk103/moneycat","last_updated":"2026-05-11"},{"id":"ccchot-osk103/pai_qwen_21102568","name":"Pai Qwen 21102568","created_by":"ccchot-osk103","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ccchot-osk103/pai_qwen_21102568","last_updated":"2026-05-11"},{"id":"charlesmccarthy/addwatermark","name":"Addwatermark","created_by":"charlesmccarthy","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/charlesmccarthy/addwatermark","last_updated":"2026-05-11"},{"id":"cjwbw/anything-v4.0","name":"Anything V4.0","created_by":"cjwbw","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/cjwbw/anything-v4.0","last_updated":"2026-05-11"},{"id":"cjwbw/semantic-segment-anything","name":"Semantic Segment Anything","created_by":"cjwbw","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/cjwbw/semantic-segment-anything","last_updated":"2026-05-11"},{"id":"codeplugtech/face-swap","name":"Face Swap","created_by":"codeplugtech","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/codeplugtech/face-swap","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"colinhughes2121/ai-anime-villain","name":"Ai Anime Villain","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-anime-villain","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-charcoal-portrait","name":"Ai Charcoal Portrait","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-charcoal-portrait","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-pop-art-stylizer","name":"Ai Pop Art Stylizer","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-pop-art-stylizer","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-tarot-card-stylizer","name":"Ai Tarot Card Stylizer","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-tarot-card-stylizer","last_updated":"2026-06-03"},{"id":"colinhughes2121/ai-vintage-film-photo","name":"Ai Vintage Film Photo","created_by":"colinhughes2121","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/colinhughes2121/ai-vintage-film-photo","last_updated":"2026-06-03"},{"id":"csviverdeia/locateanything-3b-h100","name":"Locateanything 3b H100","created_by":"csviverdeia","source":"official","status":"active","model_type":"other","parameters":3,"page_url":"https://replicate.com/csviverdeia/locateanything-3b-h100","last_updated":"2026-06-29"},{"id":"daanelson/imagebind","name":"Imagebind","created_by":"daanelson","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/daanelson/imagebind","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"damdam775/portraits_dialogues","name":"Portraits Dialogues","created_by":"damdam775","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/damdam775/portraits_dialogues","last_updated":"2026-05-11"},{"id":"dannyboy2323/character-daddanny23","name":"Character Daddanny23","created_by":"dannyboy2323","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/dannyboy2323/character-daddanny23","last_updated":"2026-06-10"},{"id":"dannyboy2323/character-danielbull23","name":"Character Danielbull23","created_by":"dannyboy2323","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/dannyboy2323/character-danielbull23","last_updated":"2026-06-10"},{"id":"dannyboy2323/character-hater","name":"Character Hater","created_by":"dannyboy2323","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/dannyboy2323/character-hater","last_updated":"2026-06-10"},{"id":"datacte/proteus-v0.2","name":"Proteus V0.2","created_by":"datacte","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/datacte/proteus-v0.2","last_updated":"2026-05-11"},{"id":"datacte/proteus-v0.3","name":"Proteus V0.3","created_by":"datacte","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/datacte/proteus-v0.3","last_updated":"2026-05-11"},{"id":"datalab-to/marker","name":"Marker","created_by":"datalab-to","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/datalab-to/marker","last_updated":"2026-05-11"},{"id":"elevenlabs/music","name":"Music","created_by":"elevenlabs","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/elevenlabs/music","last_updated":"2026-05-11"},{"id":"elevenlabs/scribe-v2","name":"Scribe V2","created_by":"elevenlabs","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/elevenlabs/scribe-v2","last_updated":"2026-05-11"},{"id":"falcons-ai/nsfw_image_detection","name":"Nsfw Image Detection","created_by":"falcons-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/falcons-ai/nsfw_image_detection","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fictions-ai/autocaption","name":"Autocaption","created_by":"fictions-ai","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fictions-ai/autocaption","last_updated":"2026-05-11"},{"id":"fishwowater/trellis2","name":"Trellis2","created_by":"fishwowater","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fishwowater/trellis2","last_updated":"2026-05-11"},{"id":"flux-kontext-apps/restore-image","name":"Restore Image","created_by":"flux-kontext-apps","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/flux-kontext-apps/restore-image","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fofr/expression-editor","name":"Expression Editor","created_by":"fofr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fofr/expression-editor","last_updated":"2026-06-10"},{"id":"fofr/face-swap-with-ideogram","name":"Face Swap With Ideogram","created_by":"fofr","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/fofr/face-swap-with-ideogram","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fofr/flux-mona-lisa","name":"Flux Mona Lisa","created_by":"fofr","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/fofr/flux-mona-lisa","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"fofr/latent-consistency-model","name":"Latent Consistency Model","created_by":"fofr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fofr/latent-consistency-model","last_updated":"2026-05-11"},{"id":"fofr/prompt-classifier","name":"Prompt Classifier","created_by":"fofr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/fofr/prompt-classifier","last_updated":"2026-06-03"},{"id":"fofr/sdxl-emoji","name":"Sdxl Emoji","created_by":"fofr","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/fofr/sdxl-emoji","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"furkkurt/vector-blog-thumbnails","name":"Vector Blog Thumbnails","created_by":"furkkurt","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/furkkurt/vector-blog-thumbnails","last_updated":"2026-05-11"},{"id":"google/gemini-3-flash","name":"Gemini 3 Flash","created_by":"google","source":"official","family":"gemini-3","status":"active","model_type":"chat","page_url":"https://replicate.com/google/gemini-3-flash","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemini-3.1-flash-tts","name":"Gemini 3.1 Flash Tts","created_by":"google","source":"official","family":"gemini-3.1","status":"active","model_type":"audio","page_url":"https://replicate.com/google/gemini-3.1-flash-tts","last_updated":"2026-05-11"},{"id":"google/gemini-3.1-pro","name":"Gemini 3.1 Pro","created_by":"google","source":"official","family":"gemini-3.1","status":"active","model_type":"chat","page_url":"https://replicate.com/google/gemini-3.1-pro","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"google/gemini-3.5-flash","name":"Gemini 3.5 Flash","created_by":"google","source":"official","family":"gemini-3.5","status":"active","model_type":"chat","page_url":"https://replicate.com/google/gemini-3.5-flash","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-20","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.5 Flash Gemini 3.5 Flash provides sustained frontier-level intelligence optimized for real-world tasks at a higher speed and lower cost. Designed for the agentic era, it excels at sub-agent deployment, multi-step workflows, and long-horizon tasks at scale. This model is particularly effective for rapid agentic loops involving complex coding cycles and iterations.","tagline":"The Interactions API is now generally available.","context_window":1048576,"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/imagen-4","name":"Imagen 4","created_by":"google","source":"official","family":"imagen-4","status":"active","model_type":"image","page_url":"https://replicate.com/google/imagen-4","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/nano-banana-2","name":"Nano Banana 2","created_by":"google","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/google/nano-banana-2","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/nano-banana","name":"Nano Banana","created_by":"google","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/google/nano-banana","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/upscaler","name":"Upscaler","created_by":"google","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/google/upscaler","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"google/veo-3.1-lite","name":"Veo 3.1 Lite","created_by":"google","source":"official","family":"veo-3.1","status":"active","model_type":"video","page_url":"https://replicate.com/google/veo-3.1-lite","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"hyper3d/rodin","name":"Rodin","created_by":"hyper3d","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/hyper3d/rodin","last_updated":"2026-05-11"},{"id":"ideogram-ai/ideogram-character","name":"Ideogram Character","created_by":"ideogram-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/ideogram-ai/ideogram-character","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"ideogram-ai/ideogram-v4-balanced","name":"Ideogram V4 Balanced","created_by":"ideogram-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/ideogram-ai/ideogram-v4-balanced","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-10"},{"id":"ideogram-ai/ideogram-v4-quality","name":"Ideogram V4 Quality","created_by":"ideogram-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/ideogram-ai/ideogram-v4-quality","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-10"},{"id":"inworld/realtime-tts-1.5-max","name":"Realtime Tts 1.5 Max","created_by":"inworld","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/inworld/realtime-tts-1.5-max","last_updated":"2026-05-11"},{"id":"inworld/realtime-tts-1.5-mini","name":"Realtime Tts 1.5 Mini","created_by":"inworld","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/inworld/realtime-tts-1.5-mini","last_updated":"2026-05-11"},{"id":"inworld/realtime-tts-2","name":"Realtime Tts 2","created_by":"inworld","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/inworld/realtime-tts-2","last_updated":"2026-05-11"},{"id":"jaaari/kokoro-82m","name":"Kokoro 82m","created_by":"jaaari","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/jaaari/kokoro-82m","last_updated":"2026-05-11"},{"id":"jaweii/hisam","name":"Hisam","created_by":"jaweii","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/jaweii/hisam","last_updated":"2026-06-29"},{"id":"jilin-dbgr/icetea","name":"Icetea","created_by":"jilin-dbgr","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/jilin-dbgr/icetea","last_updated":"2026-06-20"},{"id":"krea/krea-2-large","name":"Krea 2 Large","created_by":"krea","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/krea/krea-2-large","last_updated":"2026-06-03"},{"id":"krea/krea-2-medium","name":"Krea 2 Medium","created_by":"krea","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/krea/krea-2-medium","last_updated":"2026-06-03"},{"id":"krthr/clip-embeddings","name":"Clip Embeddings","created_by":"krthr","source":"official","status":"active","model_type":"embed","page_url":"https://replicate.com/krthr/clip-embeddings","modalities":{"input":["text"],"output":["text"]},"endpoints":["embeddings"],"last_updated":"2026-06-18"},{"id":"kwaivgi/kling-v3-omni-video","name":"Kling V3 Omni Video","created_by":"kwaivgi","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/kwaivgi/kling-v3-omni-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"lucataco/deepseek-ocr","name":"Deepseek Ocr","created_by":"lucataco","source":"official","family":"deepseek","status":"active","model_type":"chat","page_url":"https://replicate.com/lucataco/deepseek-ocr","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"lucataco/gpt-oss-safeguard-20b","name":"Gpt Oss Safeguard 20b","created_by":"lucataco","source":"official","family":"gpt-oss","status":"active","model_type":"chat","parameters":20,"page_url":"https://replicate.com/lucataco/gpt-oss-safeguard-20b","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"lucataco/kontext-realearth","name":"Kontext Realearth","created_by":"lucataco","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/lucataco/kontext-realearth","last_updated":"2026-05-11"},{"id":"lucataco/motif-video","name":"Motif Video","created_by":"lucataco","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/lucataco/motif-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"lucataco/remove-bg","name":"Remove Bg","created_by":"lucataco","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/lucataco/remove-bg","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"luma/modify-video","name":"Modify Video","created_by":"luma","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/luma/modify-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"luma/reframe-video","name":"Reframe Video","created_by":"luma","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/luma/reframe-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"lxcxsito/stargirlz","name":"Stargirlz","created_by":"lxcxsito","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/lxcxsito/stargirlz","last_updated":"2026-06-18"},{"id":"m1guelpf/nsfw-filter","name":"Nsfw Filter","created_by":"m1guelpf","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/m1guelpf/nsfw-filter","last_updated":"2026-05-11"},{"id":"maikocode/ascii-style","name":"Ascii Style","created_by":"maikocode","source":"official","status":"active","model_type":"code","page_url":"https://replicate.com/maikocode/ascii-style","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"megvii-research/nafnet","name":"Nafnet","created_by":"megvii-research","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/megvii-research/nafnet","last_updated":"2026-05-11"},{"id":"meta/llama-guard-4-12b","name":"Llama Guard 4 12b","created_by":"meta","source":"official","family":"llama-guard","status":"active","model_type":"chat","parameters":12,"page_url":"https://replicate.com/meta/llama-guard-4-12b","capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","license":"llama4","open_weight":true},{"id":"meta/sam-2-video","name":"Sam 2 Video","created_by":"meta","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/meta/sam-2-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"miike-ai/flux-ico","name":"Flux Ico","created_by":"miike-ai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/miike-ai/flux-ico","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"minimax/music-1.5","name":"Music 1.5","created_by":"minimax","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/minimax/music-1.5","last_updated":"2026-05-11"},{"id":"minimax/music-2.5","name":"Music 2.5","created_by":"minimax","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/minimax/music-2.5","last_updated":"2026-05-11"},{"id":"minimax/music-2.6","name":"Music 2.6","created_by":"minimax","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/minimax/music-2.6","last_updated":"2026-05-11"},{"id":"mptamilselvan/download-media","name":"Download Media","created_by":"mptamilselvan","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/mptamilselvan/download-media","last_updated":"2026-05-11"},{"id":"mptamilselvan/text-to-voice","name":"Text To Voice","created_by":"mptamilselvan","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/mptamilselvan/text-to-voice","last_updated":"2026-05-11"},{"id":"myaiteam2/website-scrapper","name":"Website Scrapper","created_by":"myaiteam2","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/myaiteam2/website-scrapper","last_updated":"2026-06-03"},{"id":"naimish-gami/face-swapper","name":"Face Swapper","created_by":"naimish-gami","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/naimish-gami/face-swapper","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-18"},{"id":"nicegen-ai/nsfw-filter-for-portrait","name":"Nsfw Filter For Portrait","created_by":"nicegen-ai","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/nicegen-ai/nsfw-filter-for-portrait","last_updated":"2026-06-18"},{"id":"oitoito/depth-pro","name":"Depth Pro","created_by":"oitoito","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/oitoito/depth-pro","last_updated":"2026-06-10"},{"id":"oitoito/mask2former","name":"Mask2former","created_by":"oitoito","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/oitoito/mask2former","last_updated":"2026-06-10"},{"id":"openai/gpt-4o-mini","name":"Gpt 4o Mini","created_by":"openai","source":"official","family":"gpt-4o","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-4o-mini","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-4o mini (“o” for “omni”) is a fast, affordable small model for focused tasks.","tagline":"Fast, affordable small model for focused tasks","context_window":128000,"max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-transcribe","name":"Gpt 4o Transcribe","created_by":"openai","source":"official","family":"gpt-4o","status":"active","model_type":"transcription","page_url":"https://replicate.com/openai/gpt-4o-transcribe","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11","description":"GPT-4o Transcribe is a speech-to-text model that uses GPT-4o to transcribe audio.","tagline":"Speech-to-text model powered by GPT-4o","context_window":16000,"max_output_tokens":2000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-nano","name":"Gpt 5 Nano","created_by":"openai","source":"official","family":"gpt-5","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-5-nano","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for","tagline":"Fastest, most cost-efficient version of GPT-5","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2","name":"Gpt 5.2","created_by":"openai","source":"official","family":"gpt-5.2","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-5.2","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5","name":"Gpt 5","created_by":"openai","source":"official","family":"gpt-5","status":"active","model_type":"chat","page_url":"https://replicate.com/openai/gpt-5","capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"fine_tuning":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11","description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","context_window":400000,"max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-image-1.5","name":"Gpt Image 1.5","created_by":"openai","source":"official","family":"gpt-image-1.5","status":"active","model_type":"image","page_url":"https://replicate.com/openai/gpt-image-1.5","capabilities":{"vision":true,"streaming":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"GPT Image 1.5 is our previous image generation model, with better instruction following and adherence to prompts. Learn more in our [GPT Image 1.5 usage guide](/api/docs/guides/image-generation).","tagline":"Our previous image generation model","license":"proprietary","open_weight":false},{"id":"openai/gpt-image-2","name":"Gpt Image 2","created_by":"openai","source":"official","family":"gpt-image-2","status":"active","model_type":"image","page_url":"https://replicate.com/openai/gpt-image-2","capabilities":{"vision":true,"streaming":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","description":"GPT Image 2 is our state-of-the-art image generation model for fast, high-quality image generation and editing. It supports flexible image sizes and high-fidelity image inputs. Learn more in our [image generation guide](/api/docs/guides/image-generation), or see the [pricing page](/api/docs/pricing#image-generation) and [image generation calculator](/api/docs/guides/image-generation#calculating-costs) for cost estimates.","tagline":"State-of-the-art image generation model","license":"proprietary","open_weight":false},{"id":"philz1337x/clarity-pro-upscaler","name":"Clarity Pro Upscaler","created_by":"philz1337x","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/philz1337x/clarity-pro-upscaler","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"philz1337x/crystal-video-upscaler","name":"Crystal Video Upscaler","created_by":"philz1337x","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/philz1337x/crystal-video-upscaler","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"piddnad/ddcolor","name":"Ddcolor","created_by":"piddnad","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/piddnad/ddcolor","last_updated":"2026-05-11"},{"id":"prunaai/p-image-edit","name":"P Image Edit","created_by":"prunaai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/prunaai/p-image-edit","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-06-03"},{"id":"prunaai/p-image-upscale","name":"P Image Upscale","created_by":"prunaai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/prunaai/p-image-upscale","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"prunaai/p-video-animate","name":"P Video Animate","created_by":"prunaai","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/prunaai/p-video-animate","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-18"},{"id":"prunaai/p-video-avatar","name":"P Video Avatar","created_by":"prunaai","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/prunaai/p-video-avatar","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"prunaai/p-video","name":"P Video","created_by":"prunaai","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/prunaai/p-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"prunaai/z-image-turbo","name":"Z Image Turbo","created_by":"prunaai","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/prunaai/z-image-turbo","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"qwen/qwen3-7-plus","name":"Qwen3 7 Plus","created_by":"alibaba","source":"official","family":"qwen3","status":"active","model_type":"other","page_url":"https://replicate.com/qwen/qwen3-7-plus","last_updated":"2026-06-29"},{"id":"qwen/qwen3-tts","name":"Qwen3 Tts","created_by":"alibaba","source":"official","family":"qwen3","status":"active","model_type":"audio","page_url":"https://replicate.com/qwen/qwen3-tts","last_updated":"2026-05-11"},{"id":"rafaelgalle/whisper-diarization-advanced","name":"Whisper Diarization Advanced","created_by":"rafaelgalle","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/rafaelgalle/whisper-diarization-advanced","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"replicate/train-rvc-model","name":"Train Rvc Model","created_by":"replicate","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/replicate/train-rvc-model","last_updated":"2026-05-11"},{"id":"runwayml/aleph-2","name":"Aleph 2","created_by":"runwayml","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/runwayml/aleph-2","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-10"},{"id":"runwayml/gen-4.5","name":"Gen 4.5","created_by":"runwayml","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/runwayml/gen-4.5","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"runwayml/gen4-image","name":"Gen4 Image","created_by":"runwayml","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/runwayml/gen4-image","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"schananas/grounded_sam","name":"Grounded Sam","created_by":"schananas","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/schananas/grounded_sam","last_updated":"2026-05-11"},{"id":"shreejalmaharjan-27/tiktok-short-captions","name":"Tiktok Short Captions","created_by":"shreejalmaharjan-27","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/shreejalmaharjan-27/tiktok-short-captions","last_updated":"2026-05-11"},{"id":"smokeyang/transition_generator","name":"Transition Generator","created_by":"smokeyang","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/smokeyang/transition_generator","last_updated":"2026-06-18"},{"id":"sourceful/riverflow-v2.5-pro","name":"Riverflow V2.5 Pro","created_by":"sourceful","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sourceful/riverflow-v2.5-pro","last_updated":"2026-06-18"},{"id":"sprited/birefnet-toonout","name":"Birefnet Toonout","created_by":"sprited","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sprited/birefnet-toonout","last_updated":"2026-06-18"},{"id":"sprited/birefnet-video","name":"Birefnet Video","created_by":"sprited","source":"official","status":"active","model_type":"video","page_url":"https://replicate.com/sprited/birefnet-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-18"},{"id":"sprited/birefnet","name":"Birefnet","created_by":"sprited","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sprited/birefnet","last_updated":"2026-06-18"},{"id":"spuuntries/dfk3-queeree","name":"Dfk3 Queeree","created_by":"spuuntries","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/spuuntries/dfk3-queeree","last_updated":"2026-05-11"},{"id":"sync/lipsync-2-pro","name":"Lipsync 2 Pro","created_by":"sync","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/sync/lipsync-2-pro","last_updated":"2026-05-11"},{"id":"syncodeofficial/noah-biblical-craft","name":"Noah Biblical Craft","created_by":"syncodeofficial","source":"official","status":"active","model_type":"code","page_url":"https://replicate.com/syncodeofficial/noah-biblical-craft","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"tattzy25/heavy_hand_dark","name":"Heavy Hand Dark","created_by":"tattzy25","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/tattzy25/heavy_hand_dark","last_updated":"2026-06-18"},{"id":"tencent/hunyuan-3d-3.1","name":"Hunyuan 3d 3.1","created_by":"tencent","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/tencent/hunyuan-3d-3.1","last_updated":"2026-05-11"},{"id":"thomasmol/whisper-diarization","name":"Whisper Diarization","created_by":"thomasmol","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/thomasmol/whisper-diarization","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"topazlabs/video-upscale","name":"Video Upscale","created_by":"topazlabs","source":"official","status":"active","model_type":"image","page_url":"https://replicate.com/topazlabs/video-upscale","modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11"},{"id":"ultralytics/yolo26-cls","name":"Yolo26 Cls","created_by":"ultralytics","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ultralytics/yolo26-cls","last_updated":"2026-06-29"},{"id":"ultralytics/yolo26-obb","name":"Yolo26 Obb","created_by":"ultralytics","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ultralytics/yolo26-obb","last_updated":"2026-06-29"},{"id":"ultralytics/yolo26-pose","name":"Yolo26 Pose","created_by":"ultralytics","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ultralytics/yolo26-pose","last_updated":"2026-06-29"},{"id":"ultralytics/yolo26-seg","name":"Yolo26 Seg","created_by":"ultralytics","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ultralytics/yolo26-seg","last_updated":"2026-06-29"},{"id":"ultralytics/yolo26-sem","name":"Yolo26 Sem","created_by":"ultralytics","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ultralytics/yolo26-sem","last_updated":"2026-06-29"},{"id":"ultralytics/yolo26","name":"Yolo26","created_by":"ultralytics","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/ultralytics/yolo26","last_updated":"2026-06-29"},{"id":"vaibhavs10/incredibly-fast-whisper","name":"Incredibly Fast Whisper","created_by":"vaibhavs10","source":"official","status":"active","model_type":"transcription","page_url":"https://replicate.com/vaibhavs10/incredibly-fast-whisper","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"valllllex/cartoonia_3d","name":"Cartoonia 3d","created_by":"valllllex","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/valllllex/cartoonia_3d","last_updated":"2026-05-11"},{"id":"veed/fabric-1.0","name":"Fabric 1.0","created_by":"veed","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/veed/fabric-1.0","last_updated":"2026-06-29"},{"id":"victor-upmeet/whisperx","name":"Whisperx","created_by":"victor-upmeet","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/victor-upmeet/whisperx","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-05-11"},{"id":"vm6eji6m4/whisper-chinese-pro","name":"Whisper Chinese Pro","created_by":"vm6eji6m4","source":"official","family":"whisper","status":"active","model_type":"transcription","page_url":"https://replicate.com/vm6eji6m4/whisper-chinese-pro","modalities":{"input":["audio"],"output":["text"]},"endpoints":["audio_transcriptions"],"last_updated":"2026-06-03"},{"id":"vufinder/sam3","name":"Sam3","created_by":"vufinder","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/vufinder/sam3","last_updated":"2026-06-18"},{"id":"wan-video/wan-2.2-t2v-fast","name":"Wan 2.2 T2v Fast","created_by":"wan-video","source":"official","family":"wan","status":"active","model_type":"video","page_url":"https://replicate.com/wan-video/wan-2.2-t2v-fast","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"wan-video/wan-2.5-i2v-fast","name":"Wan 2.5 I2v Fast","created_by":"wan-video","source":"official","family":"wan","status":"active","model_type":"video","page_url":"https://replicate.com/wan-video/wan-2.5-i2v-fast","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"wan-video/wan-2.5-t2v-fast","name":"Wan 2.5 T2v Fast","created_by":"wan-video","source":"official","family":"wan","status":"active","model_type":"video","page_url":"https://replicate.com/wan-video/wan-2.5-t2v-fast","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"xai/grok-4","name":"Grok 4","created_by":"xai","source":"official","family":"grok-4","status":"active","model_type":"chat","page_url":"https://replicate.com/xai/grok-4","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-05-11"},{"id":"xai/grok-imagine-image-quality","name":"Grok Imagine Image Quality","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"image","page_url":"https://replicate.com/xai/grok-imagine-image-quality","capabilities":{"vision":true,"streaming":true,"batch":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false},{"id":"xai/grok-imagine-video-1.5","name":"Grok Imagine Video 1.5","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"video","page_url":"https://replicate.com/xai/grok-imagine-video-1.5","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-06-03","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"batch":true,"vision":true}},{"id":"xai/grok-imagine-video-extension","name":"Grok Imagine Video Extension","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"video","page_url":"https://replicate.com/xai/grok-imagine-video-extension","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11"},{"id":"xai/grok-imagine-video","name":"Grok Imagine Video","created_by":"xai","source":"official","family":"grok-imagine","status":"active","model_type":"video","page_url":"https://replicate.com/xai/grok-imagine-video","modalities":{"input":["text"],"output":["video"]},"endpoints":["video"],"last_updated":"2026-05-11","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true}},{"id":"yangzhou0011/layoutgrok-qwen3-vl-cp500","name":"Layoutgrok Qwen3 Vl Cp500","created_by":"yangzhou0011","source":"official","status":"active","model_type":"chat","page_url":"https://replicate.com/yangzhou0011/layoutgrok-qwen3-vl-cp500","capabilities":{"vision":true,"streaming":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"last_updated":"2026-06-18"},{"id":"zsxkib/create-rvc-dataset","name":"Create Rvc Dataset","created_by":"zsxkib","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/zsxkib/create-rvc-dataset","last_updated":"2026-05-11"},{"id":"zsxkib/multitalk","name":"Multitalk","created_by":"zsxkib","source":"official","status":"active","model_type":"other","page_url":"https://replicate.com/zsxkib/multitalk","last_updated":"2026-05-11"},{"id":"zsxkib/realistic-voice-cloning","name":"Realistic Voice Cloning","created_by":"zsxkib","source":"official","status":"active","model_type":"audio","page_url":"https://replicate.com/zsxkib/realistic-voice-cloning","last_updated":"2026-05-11"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/sambanova.ts b/packages/npm/src/providers/sambanova.ts
index 81b3836fd..d79ea74d6 100644
--- a/packages/npm/src/providers/sambanova.ts
+++ b/packages/npm/src/providers/sambanova.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"sambanova","name":"SambaNova Cloud","region":"US","headquarters":"Palo Alto, CA","founded":2017,"url":"https://sambanova.ai","api_url":"https://api.sambanova.ai/v1","docs_url":"https://docs.sambanova.ai","pricing_url":"https://cloud.sambanova.ai/plans/pricing","description":"Inference cloud built on custom RDU dataflow chips, delivering high-throughput hosting of open-weight models.","type":"cloud","playground_url":"https://cloud.sambanova.ai/playground","openai_compatible":true,"free_tier":true,"models_url":"https://cloud.sambanova.ai/models","twitter_url":"https://x.com/SambaNovaAI","blog_url":"https://sambanova.ai/blog","github_url":"https://github.com/sambanova","icon":"","models":[{"id":"DeepSeek-R1-Distill-Llama-70B","name":"DeepSeek R1 Distill Llama 70B","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-r1-distill","description":"DeepSeek R1 reasoning distilled into a Llama 70B base.","status":"active","model_type":"reasoning","context_window":32768,"reasoning_tokens":true,"open_weight":true,"license":"llama-3","capabilities":{"reasoning":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"pricing":{"input":0.7,"output":1.4}},{"id":"DeepSeek-V3.1-cb","name":"DeepSeek V3.1 (continuous batching)","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-v3","description":"High-volume continuous-batching variant of DeepSeek V3.1 with reduced context and lower price.","status":"active","model_type":"reasoning","context_window":32768,"max_output_tokens":32768,"reasoning_tokens":true,"open_weight":true,"license":"deepseek","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.15,"output":0.75},"tagline":"High-volume continuous-batching variant of DeepSeek V3.1 with reduced context and lower price."},{"id":"DeepSeek-V3.1","name":"DeepSeek V3.1","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-v3","description":"Hybrid reasoning MoE model with switchable thinking and non-thinking modes.","status":"active","model_type":"reasoning","context_window":131072,"max_output_tokens":7168,"reasoning_tokens":true,"open_weight":true,"license":"deepseek","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":3,"output":4.5},"tagline":"Hybrid reasoning MoE model with switchable thinking and non-thinking modes."},{"id":"DeepSeek-V3.2","name":"DeepSeek V3.2","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-v3","description":"Preview release of DeepSeek V3.2 on SambaCloud, high volume lane with limited context.","status":"preview","model_type":"reasoning","context_window":32768,"max_output_tokens":7168,"reasoning_tokens":true,"open_weight":true,"license":"deepseek","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":3,"output":4.5},"tagline":"Preview release of DeepSeek V3.2 on SambaCloud, high volume lane with limited context."},{"id":"gemma-3-12b-it","name":"Gemma 3 12B IT","created_by":"google","source":"official","last_updated":"2026-06-03","family":"gemma-3","description":"Google Gemma 3 12B instruction-tuned open-weights model.","status":"active","model_type":"chat","context_window":131072,"max_output_tokens":131072,"open_weight":true,"license":"gemma","capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"pricing":{"input":0.35,"output":0.59},"tagline":"Google Gemma 3 12B instruction-tuned open-weights model."},{"id":"gemma-4-31B-it","name":"gemma-4-31B-it","created_by":"google","source":"official","family":"gemma-4","status":"active","context_window":131072,"max_output_tokens":131072,"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.38,"output":1.15},"endpoints":["chat_completions"],"last_updated":"2026-06-03"},{"id":"gpt-oss-120b","name":"GPT-OSS 120B","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-oss","description":"OpenAI's open-weight GPT-OSS 120B reasoning model.","status":"active","model_type":"reasoning","context_window":131072,"max_output_tokens":131072,"reasoning_tokens":true,"open_weight":true,"license":"apache-2.0","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"vision":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.22,"output":0.59},"tagline":"OpenAI's open-weight GPT-OSS 120B reasoning model.","knowledge_cutoff":"2024-06","parameters":120},{"id":"Llama-4-Maverick-17B-128E-Instruct","name":"Llama 4 Maverick 17B 128E Instruct","created_by":"meta","source":"official","last_updated":"2026-04-26","family":"llama-4","description":"Llama 4 Maverick MoE multimodal model (17B active, 128 experts) with vision input.","status":"preview","model_type":"chat","context_window":131072,"max_output_tokens":4096,"open_weight":true,"license":"llama-4","capabilities":{"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.63,"output":1.8},"tagline":"Llama 4 Maverick MoE multimodal model (17B active, 128 experts) with vision input.","parameters":402},{"id":"Meta-Llama-3.3-70B-Instruct","name":"Llama 3.3 70B Instruct","created_by":"meta","source":"official","last_updated":"2026-04-26","family":"llama-3","description":"Meta's Llama 3.3 70B instruction-tuned model.","status":"active","model_type":"chat","context_window":131072,"max_output_tokens":3072,"open_weight":true,"license":"llama-3.3","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.6,"output":1.2},"tagline":"Meta's Llama 3.3 70B instruction-tuned model."},{"id":"MiniMax-M2.5","name":"MiniMax M2.5","created_by":"minimax","source":"official","last_updated":"2026-04-26","family":"minimax-m2","description":"MiniMax M2.5 agentic and coding-focused open-weights model with long context.","status":"active","model_type":"chat","context_window":163840,"max_output_tokens":163840,"open_weight":true,"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.3,"output":1.2},"tagline":"MiniMax M2.5 agentic and coding-focused open-weights model with long context.","license":"mit"},{"id":"MiniMax-M2.7","name":"MiniMax-M2.7","created_by":"minimax","source":"official","family":"minimax","status":"active","context_window":196608,"max_output_tokens":196608,"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.4},"endpoints":["chat_completions"],"last_updated":"2026-06-03","description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true}}]};
+export const provider: ProviderWithModels = {"id":"sambanova","name":"SambaNova Cloud","region":"US","headquarters":"Palo Alto, CA","founded":2017,"url":"https://sambanova.ai","api_url":"https://api.sambanova.ai/v1","docs_url":"https://docs.sambanova.ai","pricing_url":"https://cloud.sambanova.ai/plans/pricing","description":"Inference cloud built on custom RDU dataflow chips, delivering high-throughput hosting of open-weight models.","type":"cloud","playground_url":"https://cloud.sambanova.ai/playground","openai_compatible":true,"free_tier":true,"models_url":"https://cloud.sambanova.ai/models","twitter_url":"https://x.com/SambaNovaAI","blog_url":"https://sambanova.ai/blog","github_url":"https://github.com/sambanova","icon":"","models":[{"id":"DeepSeek-R1-Distill-Llama-70B","name":"DeepSeek R1 Distill Llama 70B","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-r1-distill","description":"DeepSeek R1 reasoning distilled into a Llama 70B base.","status":"active","model_type":"reasoning","context_window":32768,"reasoning_tokens":true,"open_weight":true,"license":"llama-3","capabilities":{"reasoning":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"pricing":{"input":0.7,"output":1.4}},{"id":"DeepSeek-V3.1-cb","name":"DeepSeek V3.1 (continuous batching)","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-v3","description":"High-volume continuous-batching variant of DeepSeek V3.1 with reduced context and lower price.","status":"active","model_type":"reasoning","context_window":32768,"max_output_tokens":32768,"reasoning_tokens":true,"open_weight":true,"license":"deepseek","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.15,"output":0.75},"tagline":"High-volume continuous-batching variant of DeepSeek V3.1 with reduced context and lower price."},{"id":"DeepSeek-V3.1","name":"DeepSeek V3.1","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-v3","description":"Hybrid reasoning MoE model with switchable thinking and non-thinking modes.","status":"active","model_type":"reasoning","context_window":131072,"max_output_tokens":7168,"reasoning_tokens":true,"open_weight":true,"license":"deepseek","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":3,"output":4.5},"tagline":"Hybrid reasoning MoE model with switchable thinking and non-thinking modes."},{"id":"DeepSeek-V3.2","name":"DeepSeek V3.2","created_by":"deepseek","source":"official","last_updated":"2026-04-26","family":"deepseek-v3","description":"Preview release of DeepSeek V3.2 on SambaCloud, high volume lane with limited context.","status":"preview","model_type":"reasoning","context_window":32768,"max_output_tokens":7168,"reasoning_tokens":true,"open_weight":true,"license":"deepseek","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":3,"output":4.5},"tagline":"Preview release of DeepSeek V3.2 on SambaCloud, high volume lane with limited context."},{"id":"gemma-3-12b-it","name":"Gemma 3 12B IT","created_by":"google","source":"official","last_updated":"2026-06-03","family":"gemma-3","description":"Google Gemma 3 12B instruction-tuned open-weights model.","status":"active","model_type":"chat","context_window":131072,"max_output_tokens":131072,"open_weight":true,"license":"gemma","capabilities":{"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"pricing":{"input":0.35,"output":0.59},"tagline":"Google Gemma 3 12B instruction-tuned open-weights model."},{"id":"gemma-4-31B-it","name":"gemma-4-31B-it","created_by":"google","source":"official","family":"gemma-4","status":"active","context_window":131072,"max_output_tokens":131072,"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.22,"output":0.59},"endpoints":["chat_completions"],"last_updated":"2026-06-29"},{"id":"gpt-oss-120b","name":"GPT-OSS 120B","created_by":"openai","source":"official","last_updated":"2026-04-26","family":"gpt-oss","description":"OpenAI's open-weight GPT-OSS 120B reasoning model.","status":"active","model_type":"reasoning","context_window":131072,"max_output_tokens":131072,"reasoning_tokens":true,"open_weight":true,"license":"apache-2.0","capabilities":{"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"streaming":true,"fine_tuning":true,"vision":false,"batch":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.22,"output":0.59},"tagline":"OpenAI's open-weight GPT-OSS 120B reasoning model.","knowledge_cutoff":"2024-06","parameters":120},{"id":"Llama-4-Maverick-17B-128E-Instruct","name":"Llama 4 Maverick 17B 128E Instruct","created_by":"meta","source":"official","last_updated":"2026-04-26","family":"llama-4","description":"Llama 4 Maverick MoE multimodal model (17B active, 128 experts) with vision input.","status":"preview","model_type":"chat","context_window":131072,"max_output_tokens":4096,"open_weight":true,"license":"llama-4","capabilities":{"vision":true,"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.63,"output":1.8},"tagline":"Llama 4 Maverick MoE multimodal model (17B active, 128 experts) with vision input.","parameters":402},{"id":"Meta-Llama-3.3-70B-Instruct","name":"Llama 3.3 70B Instruct","created_by":"meta","source":"official","last_updated":"2026-04-26","family":"llama-3","description":"Meta's Llama 3.3 70B instruction-tuned model.","status":"active","model_type":"chat","context_window":131072,"max_output_tokens":3072,"open_weight":true,"license":"llama-3.3","capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.6,"output":1.2},"tagline":"Meta's Llama 3.3 70B instruction-tuned model."},{"id":"MiniMax-M2.5","name":"MiniMax M2.5","created_by":"minimax","source":"official","last_updated":"2026-04-26","family":"minimax-m2","description":"MiniMax M2.5 agentic and coding-focused open-weights model with long context.","status":"active","model_type":"chat","context_window":163840,"max_output_tokens":163840,"open_weight":true,"capabilities":{"tool_call":true,"structured_output":true,"json_mode":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"endpoints":["chat_completions"],"pricing":{"input":0.3,"output":1.2},"tagline":"MiniMax M2.5 agentic and coding-focused open-weights model with long context.","license":"mit"},{"id":"MiniMax-M2.7","name":"MiniMax-M2.7","created_by":"minimax","source":"official","family":"minimax","status":"active","context_window":196608,"max_output_tokens":196608,"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.6,"output":2.4},"endpoints":["chat_completions"],"last_updated":"2026-06-03","description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true}}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/stability.ts b/packages/npm/src/providers/stability.ts
index ca8e7da2a..0e3bcf321 100644
--- a/packages/npm/src/providers/stability.ts
+++ b/packages/npm/src/providers/stability.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"stability","name":"Stability AI","region":"GB","headquarters":"London, United Kingdom","founded":2019,"url":"https://stability.ai","api_url":"https://api.stability.ai","docs_url":"https://platform.stability.ai/docs","pricing_url":"https://platform.stability.ai/pricing","description":"AI company behind Stable Diffusion, Stable Video, Stable Audio, and related open and hosted generative models.","type":"direct","models_url":"https://huggingface.co/stabilityai","openai_compatible":false,"free_tier":false,"github_url":"https://github.com/Stability-AI","twitter_url":"https://x.com/StabilityAI","icon":"","models":[{"id":"ar-stablelm-2-base","name":"Ar Stablelm 2 Base","created_by":"stability","source":"official","description":"Ar Stablelm 2 Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-07-29","model_type":"chat","license":"other","page_url":"https://huggingface.co/stabilityai/ar-stablelm-2-base","open_weight":true,"tagline":"Ar Stablelm 2 Base is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":2,"last_modified":"2024-12-06","last_updated":"2026-06-03"},{"id":"ar-stablelm-2-chat-gguf","name":"Ar Stablelm 2 Chat Gguf","created_by":"stability","source":"official","description":"Ar Stablelm 2 Chat Gguf is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-08-09","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/ar-stablelm-2-chat-gguf","open_weight":true,"tagline":"Ar Stablelm 2 Chat Gguf is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":6,"last_modified":"2024-12-20","last_updated":"2026-06-03"},{"id":"ar-stablelm-2-chat","name":"Ar Stablelm 2 Chat","created_by":"stability","source":"official","description":"Ar Stablelm 2 Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-08-06","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/ar-stablelm-2-chat","open_weight":true,"tagline":"Ar Stablelm 2 Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":58,"last_modified":"2024-12-06","last_updated":"2026-06-03"},{"id":"arcade100k","name":"Arcade100k","created_by":"stability","source":"official","description":"Arcade100k is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-27","model_type":"other","page_url":"https://huggingface.co/stabilityai/arcade100k","open_weight":true,"tagline":"Arcade100k is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":0,"last_modified":"2024-04-12","last_updated":"2026-05-11"},{"id":"codellama13b_instruct_260k_synthesis","name":"Codellama13b Instruct 260k Synthesis","created_by":"stability","source":"official","family":"codellama","description":"Codellama13b Instruct 260k Synthesis is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-08","model_type":"chat","page_url":"https://huggingface.co/stabilityai/codellama13b_instruct_260k_synthesis","open_weight":true,"tagline":"Codellama13b Instruct 260k Synthesis is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":38,"last_modified":"2023-11-08","last_updated":"2026-06-20"},{"id":"control-lora","name":"Control Lora","created_by":"stability","source":"official","description":"Control Lora is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-17","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/control-lora","open_weight":true,"tagline":"Control Lora is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2023-08-19","last_updated":"2026-05-11"},{"id":"cosxl","name":"Cosxl","created_by":"stability","source":"official","description":"Cosxl is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-08","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/cosxl","open_weight":true,"tagline":"Cosxl is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2024-04-13","last_updated":"2026-05-11"},{"id":"japanese-instructblip-alpha","name":"Japanese Instructblip Alpha","created_by":"stability","source":"official","description":"Japanese Instructblip Alpha is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-15","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-instructblip-alpha","open_weight":true,"tagline":"Japanese Instructblip Alpha is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":74,"last_modified":"2023-11-17","last_updated":"2026-06-20"},{"id":"japanese-stable-clip-vit-l-16","name":"Japanese Stable Clip Vit L 16","created_by":"stability","source":"official","description":"Japanese Stable Clip Vit L 16 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-01","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-stable-clip-vit-l-16","open_weight":true,"tagline":"Japanese Stable Clip Vit L 16 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1962,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"japanese-stable-diffusion-xl","name":"Japanese Stable Diffusion Xl","created_by":"stability","source":"official","description":"Japanese Stable Diffusion Xl is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-01","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-stable-diffusion-xl","open_weight":true,"tagline":"Japanese Stable Diffusion Xl is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":48,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"japanese-stable-vlm","name":"Japanese Stable Vlm","created_by":"stability","source":"official","description":"Japanese Stable Vlm is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-01","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-stable-vlm","open_weight":true,"tagline":"Japanese Stable Vlm is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":8,"last_modified":"2024-07-10","last_updated":"2026-06-18"},{"id":"japanese-stablelm-2-base-1_6b","name":"Japanese Stablelm 2 Base 1 6b","created_by":"stability","source":"official","description":"Japanese Stablelm 2 Base 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-02","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-2-base-1_6b","open_weight":true,"tagline":"Japanese Stablelm 2 Base 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3,"last_modified":"2024-05-02","last_updated":"2026-06-03"},{"id":"japanese-stablelm-2-instruct-1_6b","name":"Japanese Stablelm 2 Instruct 1 6b","created_by":"stability","source":"official","description":"Japanese Stablelm 2 Instruct 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-02","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-2-instruct-1_6b","open_weight":true,"tagline":"Japanese Stablelm 2 Instruct 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":68,"last_modified":"2024-07-10","last_updated":"2026-06-03"},{"id":"japanese-stablelm-3b-4e1t-base","name":"Japanese Stablelm 3b 4e1t Base","created_by":"stability","source":"official","description":"Japanese Stablelm 3b 4e1t Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-base","open_weight":true,"tagline":"Japanese Stablelm 3b 4e1t Base is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":56,"last_modified":"2024-04-26","last_updated":"2026-06-20"},{"id":"japanese-stablelm-3b-4e1t-instruct","name":"Japanese Stablelm 3b 4e1t Instruct","created_by":"stability","source":"official","description":"Japanese Stablelm 3b 4e1t Instruct is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-instruct","open_weight":true,"tagline":"Japanese Stablelm 3b 4e1t Instruct is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":156,"last_modified":"2024-04-26","last_updated":"2026-06-20"},{"id":"japanese-stablelm-base-alpha-7b","name":"Japanese Stablelm Base Alpha 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-09","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-alpha-7b","open_weight":true,"tagline":"Japanese Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":661,"last_modified":"2023-08-22","last_updated":"2026-06-20"},{"id":"japanese-stablelm-base-beta-70b","name":"Japanese Stablelm Base Beta 70b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":70,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-beta-70b","open_weight":true,"tagline":"Japanese Stablelm Base Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":130,"last_modified":"2023-12-19","last_updated":"2026-06-20"},{"id":"japanese-stablelm-base-beta-7b","name":"Japanese Stablelm Base Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Base Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":53,"last_modified":"2023-12-19","last_updated":"2026-06-20"},{"id":"japanese-stablelm-base-gamma-7b","name":"Japanese Stablelm Base Gamma 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-gamma-7b","open_weight":true,"tagline":"Japanese Stablelm Base Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":144,"last_modified":"2024-01-25","last_updated":"2026-06-20"},{"id":"japanese-stablelm-base-ja_vocab-beta-7b","name":"Japanese Stablelm Base Ja Vocab Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-ja_vocab-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Base Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":38,"last_modified":"2023-12-19","last_updated":"2026-06-20"},{"id":"japanese-stablelm-instruct-alpha-7b-v2","name":"Japanese Stablelm Instruct Alpha 7b V2","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-09-26","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b-v2","open_weight":true,"tagline":"Japanese Stablelm Instruct Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":49,"last_modified":"2023-10-06","last_updated":"2026-06-20"},{"id":"japanese-stablelm-instruct-alpha-7b","name":"Japanese Stablelm Instruct Alpha 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-10","model_type":"chat","license":"other","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":0,"last_modified":"2023-08-22","last_updated":"2026-05-11"},{"id":"japanese-stablelm-instruct-beta-70b","name":"Japanese Stablelm Instruct Beta 70b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":70,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-70b","open_weight":true,"tagline":"Japanese Stablelm Instruct Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":130,"last_modified":"2023-12-19","last_updated":"2026-06-20"},{"id":"japanese-stablelm-instruct-beta-7b","name":"Japanese Stablelm Instruct Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":86,"last_modified":"2023-12-19","last_updated":"2026-06-20"},{"id":"japanese-stablelm-instruct-gamma-7b","name":"Japanese Stablelm Instruct Gamma 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-gamma-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":149,"last_modified":"2024-01-24","last_updated":"2026-06-20"},{"id":"japanese-stablelm-instruct-ja_vocab-beta-7b","name":"Japanese Stablelm Instruct Ja Vocab Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-ja_vocab-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":44,"last_modified":"2023-12-19","last_updated":"2026-06-20"},{"id":"SAME-L","name":"SAME L","created_by":"stability","source":"official","description":"SAME L is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/SAME-L","open_weight":true,"tagline":"SAME L is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3983,"last_modified":"2026-06-18","last_updated":"2026-06-20"},{"id":"SAME-S","name":"SAME S","created_by":"stability","source":"official","description":"SAME S is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/SAME-S","open_weight":true,"tagline":"SAME S is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1660,"last_modified":"2026-05-19","last_updated":"2026-06-20"},{"id":"sd-turbo","name":"Sd Turbo","created_by":"stability","source":"official","description":"Sd Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-27","model_type":"image","page_url":"https://huggingface.co/stabilityai/sd-turbo","open_weight":true,"tagline":"Sd Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":821422,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"sd-vae-ft-ema-original","name":"Sd Vae Ft Ema Original","created_by":"stability","source":"official","description":"Sd Vae Ft Ema Original is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"image","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-ema-original","open_weight":true,"tagline":"Sd Vae Ft Ema Original is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2022-12-19","last_updated":"2026-05-11"},{"id":"sd-vae-ft-ema","name":"Sd Vae Ft Ema","created_by":"stability","source":"official","description":"Sd Vae Ft Ema is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-ema","open_weight":true,"tagline":"Sd Vae Ft Ema is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":7329,"last_modified":"2023-06-05","last_updated":"2026-06-20"},{"id":"sd-vae-ft-mse-original","name":"Sd Vae Ft Mse Original","created_by":"stability","source":"official","description":"Sd Vae Ft Mse Original is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"image","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-mse-original","open_weight":true,"tagline":"Sd Vae Ft Mse Original is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2022-12-19","last_updated":"2026-05-11"},{"id":"sd-vae-ft-mse","name":"Sd Vae Ft Mse","created_by":"stability","source":"official","description":"Sd Vae Ft Mse is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-mse","open_weight":true,"tagline":"Sd Vae Ft Mse is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":108635,"last_modified":"2023-06-06","last_updated":"2026-06-20"},{"id":"sd-x2-latent-upscaler","name":"Sd X2 Latent Upscaler","created_by":"stability","source":"official","description":"Sd X2 Latent Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-02-03","model_type":"other","license":"openrail++","page_url":"https://huggingface.co/stabilityai/sd-x2-latent-upscaler","open_weight":true,"tagline":"Sd X2 Latent Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":132707,"last_modified":"2023-06-05","last_updated":"2026-06-20"},{"id":"sdxl-turbo-ryzen-ai","name":"Sdxl Turbo Ryzen Ai","created_by":"stability","source":"official","description":"Sdxl Turbo Ryzen Ai is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-12-06","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/sdxl-turbo-ryzen-ai","open_weight":true,"tagline":"Sdxl Turbo Ryzen Ai is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":638,"last_modified":"2024-12-11","last_updated":"2026-06-20"},{"id":"sdxl-turbo-tensorrt","name":"Sdxl Turbo Tensorrt","created_by":"stability","source":"official","description":"Sdxl Turbo Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-12-28","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/sdxl-turbo-tensorrt","open_weight":true,"tagline":"Sdxl Turbo Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":30,"last_modified":"2025-10-20","last_updated":"2026-06-18"},{"id":"sdxl-turbo","name":"Sdxl Turbo","created_by":"stability","source":"official","description":"Sdxl Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-27","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/sdxl-turbo","open_weight":true,"tagline":"Sdxl Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":703323,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"sdxl-vae","name":"Sdxl Vae","created_by":"stability","source":"official","description":"Sdxl Vae is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-06-21","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/sdxl-vae","open_weight":true,"tagline":"Sdxl Vae is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":276759,"last_modified":"2023-08-04","last_updated":"2026-06-20"},{"id":"sp4d","name":"Sp4d","created_by":"stability","source":"official","description":"Sp4d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-11-03","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/sp4d","open_weight":true,"tagline":"Sp4d is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":18,"last_modified":"2025-11-05","last_updated":"2026-06-20"},{"id":"stable-audio-3-medium-base","name":"Stable Audio 3 Medium Base","created_by":"stability","source":"official","description":"Stable Audio 3 Medium Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-medium-base","open_weight":true,"tagline":"Stable Audio 3 Medium Base is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":3727,"last_modified":"2026-06-16","last_updated":"2026-06-20"},{"id":"stable-audio-3-medium","name":"Stable Audio 3 Medium","created_by":"stability","source":"official","description":"Stable Audio 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-medium","open_weight":true,"tagline":"Stable Audio 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":61606,"last_modified":"2026-06-16","last_updated":"2026-06-20"},{"id":"stable-audio-3-small-music-base","name":"Stable Audio 3 Small Music Base","created_by":"stability","source":"official","description":"Stable Audio 3 Small Music Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-16","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-music-base","open_weight":true,"tagline":"Stable Audio 3 Small Music Base is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":2040,"last_modified":"2026-05-20","last_updated":"2026-06-20"},{"id":"stable-audio-3-small-music","name":"Stable Audio 3 Small Music","created_by":"stability","source":"official","description":"Stable Audio 3 Small Music is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-music","open_weight":true,"tagline":"Stable Audio 3 Small Music is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":21350,"last_modified":"2026-05-19","last_updated":"2026-06-20"},{"id":"stable-audio-3-small-sfx-base","name":"Stable Audio 3 Small Sfx Base","created_by":"stability","source":"official","description":"Stable Audio 3 Small Sfx Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-sfx-base","open_weight":true,"tagline":"Stable Audio 3 Small Sfx Base is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":1347,"last_modified":"2026-05-20","last_updated":"2026-06-20"},{"id":"stable-audio-3-small-sfx","name":"Stable Audio 3 Small Sfx","created_by":"stability","source":"official","description":"Stable Audio 3 Small Sfx is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-sfx","open_weight":true,"tagline":"Stable Audio 3 Small Sfx is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":12184,"last_modified":"2026-05-19","last_updated":"2026-06-20"},{"id":"stable-audio-open-1.0","name":"Stable Audio Open 1.0","created_by":"stability","source":"official","description":"Stable Audio Open 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-24","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-open-1.0","open_weight":true,"tagline":"Stable Audio Open 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":43500,"last_modified":"2025-06-19","last_updated":"2026-06-20"},{"id":"stable-audio-open-small","name":"Stable Audio Open Small","created_by":"stability","source":"official","description":"Stable Audio Open Small is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-05-12","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-open-small","open_weight":true,"tagline":"Stable Audio Open Small is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":2214,"last_modified":"2025-05-27","last_updated":"2026-06-20"},{"id":"stable-cascade-prior","name":"Stable Cascade Prior","created_by":"stability","source":"official","description":"Stable Cascade Prior is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-12","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-cascade-prior","open_weight":true,"tagline":"Stable Cascade Prior is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":867,"last_modified":"2024-03-14","last_updated":"2026-06-20"},{"id":"stable-cascade","name":"Stable Cascade","created_by":"stability","source":"official","description":"Stable Cascade is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-06","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-cascade","open_weight":true,"tagline":"Stable Cascade is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":8730,"last_modified":"2024-03-16","last_updated":"2026-06-20"},{"id":"stable-code-3b","name":"Stable Code 3b","created_by":"stability","source":"official","description":"Stable Code 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-01-09","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stable-code-3b","open_weight":true,"tagline":"Stable Code 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":6085,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stable-code-instruct-3b","name":"Stable Code Instruct 3b","created_by":"stability","source":"official","description":"Stable Code Instruct 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-06","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stable-code-instruct-3b","open_weight":true,"tagline":"Stable Code Instruct 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":2533,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stable-codec-speech-16k-base","name":"Stable Codec Speech 16k Base","created_by":"stability","source":"official","description":"Stable Codec Speech 16k Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-01-14","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-codec-speech-16k-base","open_weight":true,"tagline":"Stable Codec Speech 16k Base is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":1,"last_modified":"2025-01-14","last_updated":"2026-06-03"},{"id":"stable-codec-speech-16k","name":"Stable Codec Speech 16k","created_by":"stability","source":"official","description":"Stable Codec Speech 16k is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-01-10","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-codec-speech-16k","open_weight":true,"tagline":"Stable Codec Speech 16k is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":85,"last_modified":"2025-01-10","last_updated":"2026-06-20"},{"id":"stable-diffusion-3-medium-amdnpu","name":"Stable Diffusion 3 Medium Amdnpu","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Amdnpu is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-01","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium-amdnpu","open_weight":true,"tagline":"Stable Diffusion 3 Medium Amdnpu is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":39,"last_modified":"2026-05-05","last_updated":"2026-06-20"},{"id":"stable-diffusion-3-medium-diffusers","name":"Stable Diffusion 3 Medium Diffusers","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Diffusers is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-06-12","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium-diffusers","open_weight":true,"tagline":"Stable Diffusion 3 Medium Diffusers is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":57971,"last_modified":"2024-06-19","last_updated":"2026-06-20"},{"id":"stable-diffusion-3-medium-tensorrt","name":"Stable Diffusion 3 Medium Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-06-12","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":10,"last_modified":"2025-10-20","last_updated":"2026-06-20"},{"id":"stable-diffusion-3-medium","name":"Stable Diffusion 3 Medium","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-30","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium","open_weight":true,"tagline":"Stable Diffusion 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":3165,"last_modified":"2024-08-12","last_updated":"2026-06-20"},{"id":"stable-diffusion-3-medium_amdgpu","name":"Stable Diffusion 3 Medium Amdgpu","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Amdgpu is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-02-28","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium_amdgpu","open_weight":true,"tagline":"Stable Diffusion 3 Medium Amdgpu is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2025-07-16","last_updated":"2026-05-11"},{"id":"stable-diffusion-3.5-controlnets-tensorrt","name":"Stable Diffusion 3.5 Controlnets Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Controlnets Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-08-07","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-controlnets-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3.5 Controlnets Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":358,"last_modified":"2025-10-20","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-controlnets","name":"Stable Diffusion 3.5 Controlnets","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Controlnets is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-15","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-controlnets","open_weight":true,"tagline":"Stable Diffusion 3.5 Controlnets is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":11,"last_modified":"2024-12-11","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-large-controlnet-blur","name":"Stable Diffusion 3.5 Large Controlnet Blur","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Controlnet Blur is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-25","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-controlnet-blur","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Controlnet Blur is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":72,"last_modified":"2024-11-28","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-large-controlnet-canny","name":"Stable Diffusion 3.5 Large Controlnet Canny","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Controlnet Canny is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-25","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-controlnet-canny","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Controlnet Canny is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":40886,"last_modified":"2024-11-28","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-large-controlnet-depth","name":"Stable Diffusion 3.5 Large Controlnet Depth","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Controlnet Depth is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-25","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-controlnet-depth","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Controlnet Depth is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":45326,"last_modified":"2024-11-28","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-large-tensorrt","name":"Stable Diffusion 3.5 Large Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-05-16","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":2872,"last_modified":"2025-10-20","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-large-turbo","name":"Stable Diffusion 3.5 Large Turbo","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-10-22","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-turbo","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":7251,"last_modified":"2024-10-22","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-large","name":"Stable Diffusion 3.5 Large","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-10-22","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large","open_weight":true,"tagline":"Stable Diffusion 3.5 Large is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":24421,"last_modified":"2024-10-22","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-medium-tensorrt","name":"Stable Diffusion 3.5 Medium Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-08-07","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-medium-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3.5 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":18,"last_modified":"2025-10-20","last_updated":"2026-06-20"},{"id":"stable-diffusion-3.5-medium","name":"Stable Diffusion 3.5 Medium","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-10-29","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-medium","open_weight":true,"tagline":"Stable Diffusion 3.5 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":93106,"last_modified":"2024-10-31","last_updated":"2026-06-20"},{"id":"stable-diffusion-x4-upscaler","name":"Stable Diffusion X4 Upscaler","created_by":"stability","source":"official","description":"Stable Diffusion X4 Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-11-23","model_type":"other","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-x4-upscaler","open_weight":true,"tagline":"Stable Diffusion X4 Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":17651,"last_modified":"2023-07-05","last_updated":"2026-06-20"},{"id":"stable-diffusion-xl-1.0-tensorrt","name":"Stable Diffusion Xl 1.0 Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion Xl 1.0 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-22","model_type":"image","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-1.0-tensorrt","open_weight":true,"tagline":"Stable Diffusion Xl 1.0 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":176,"last_modified":"2025-10-20","last_updated":"2026-06-20"},{"id":"stable-diffusion-xl-base-0.9","name":"Stable Diffusion Xl Base 0.9","created_by":"stability","source":"official","description":"Stable Diffusion Xl Base 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-06-19","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-base-0.9","open_weight":true,"tagline":"Stable Diffusion Xl Base 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":208,"last_modified":"2023-07-12","last_updated":"2026-06-20"},{"id":"stable-diffusion-xl-base-1.0","name":"Stable Diffusion Xl Base 1.0","created_by":"stability","source":"official","description":"Stable Diffusion Xl Base 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-25","model_type":"image","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0","open_weight":true,"tagline":"Stable Diffusion Xl Base 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":1417752,"last_modified":"2023-10-30","last_updated":"2026-06-20"},{"id":"stable-diffusion-xl-refiner-0.9","name":"Stable Diffusion Xl Refiner 0.9","created_by":"stability","source":"official","description":"Stable Diffusion Xl Refiner 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-06-21","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-0.9","open_weight":true,"tagline":"Stable Diffusion Xl Refiner 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":51,"last_modified":"2023-07-12","last_updated":"2026-06-20"},{"id":"stable-diffusion-xl-refiner-1.0","name":"Stable Diffusion Xl Refiner 1.0","created_by":"stability","source":"official","description":"Stable Diffusion Xl Refiner 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-26","model_type":"image","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0","open_weight":true,"tagline":"Stable Diffusion Xl Refiner 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":175890,"last_modified":"2023-09-25","last_updated":"2026-06-20"},{"id":"stable-fast-3d","name":"Stable Fast 3d","created_by":"stability","source":"official","description":"Stable Fast 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-07-17","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-fast-3d","open_weight":true,"tagline":"Stable Fast 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"huggingface_downloads":9086,"last_modified":"2025-04-08","last_updated":"2026-06-20"},{"id":"stable-point-aware-3d","name":"Stable Point Aware 3d","created_by":"stability","source":"official","description":"Stable Point Aware 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-12-28","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-point-aware-3d","open_weight":true,"tagline":"Stable Point Aware 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"huggingface_downloads":2285,"last_modified":"2025-04-08","last_updated":"2026-06-20"},{"id":"stable-video-diffusion-img2vid-xt-1-1-tensorrt","name":"Stable Video Diffusion Img2vid Xt 1 1 Tensorrt","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid Xt 1 1 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-07","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt-1-1-tensorrt","open_weight":true,"tagline":"Stable Video Diffusion Img2vid Xt 1 1 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":43,"last_modified":"2025-10-20","last_updated":"2026-06-20"},{"id":"stable-video-diffusion-img2vid-xt-1-1","name":"Stable Video Diffusion Img2vid Xt 1 1","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid Xt 1 1 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-02","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt-1-1","open_weight":true,"tagline":"Stable Video Diffusion Img2vid Xt 1 1 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":8806,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stable-video-diffusion-img2vid-xt","name":"Stable Video Diffusion Img2vid Xt","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid Xt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-20","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt","open_weight":true,"tagline":"Stable Video Diffusion Img2vid Xt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":229371,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stable-video-diffusion-img2vid","name":"Stable Video Diffusion Img2vid","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-20","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid","open_weight":true,"tagline":"Stable Video Diffusion Img2vid is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":33139,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stable-virtual-camera","name":"Stable Virtual Camera","created_by":"stability","source":"official","description":"Stable Virtual Camera is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-02-19","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-virtual-camera","open_weight":true,"tagline":"Stable Virtual Camera is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":11482,"last_modified":"2025-06-03","last_updated":"2026-06-20"},{"id":"stable-zero123","name":"Stable Zero123","created_by":"stability","source":"official","description":"Stable Zero123 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-12-08","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-zero123","open_weight":true,"tagline":"Stable Zero123 is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":0,"last_modified":"2024-07-10","last_updated":"2026-05-11"},{"id":"StableBeluga-13B","name":"StableBeluga 13B","created_by":"stability","source":"official","description":"StableBeluga 13B is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-27","model_type":"chat","parameters":13,"page_url":"https://huggingface.co/stabilityai/StableBeluga-13B","open_weight":true,"tagline":"StableBeluga 13B is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":274,"last_modified":"2023-08-29","last_updated":"2026-06-20"},{"id":"StableBeluga-7B","name":"StableBeluga 7B","created_by":"stability","source":"official","description":"StableBeluga 7B is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-27","model_type":"chat","parameters":7,"page_url":"https://huggingface.co/stabilityai/StableBeluga-7B","open_weight":true,"tagline":"StableBeluga 7B is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":379,"last_modified":"2023-08-29","last_updated":"2026-06-20"},{"id":"StableBeluga1-Delta","name":"StableBeluga1 Delta","created_by":"stability","source":"official","description":"StableBeluga1 Delta is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-20","model_type":"chat","license":"cc-by-nc-4.0","page_url":"https://huggingface.co/stabilityai/StableBeluga1-Delta","open_weight":true,"tagline":"StableBeluga1 Delta is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":139,"last_modified":"2023-07-27","last_updated":"2026-06-20"},{"id":"StableBeluga2","name":"StableBeluga2","created_by":"stability","source":"official","description":"StableBeluga2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-20","model_type":"chat","page_url":"https://huggingface.co/stabilityai/StableBeluga2","open_weight":true,"tagline":"StableBeluga2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":752,"last_modified":"2023-09-18","last_updated":"2026-06-20"},{"id":"stablecode-completion-alpha-3b-4k","name":"Stablecode Completion Alpha 3b 4k","created_by":"stability","source":"official","description":"Stablecode Completion Alpha 3b 4k is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-07","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablecode-completion-alpha-3b-4k","open_weight":true,"tagline":"Stablecode Completion Alpha 3b 4k is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1116,"last_modified":"2023-08-08","last_updated":"2026-06-20"},{"id":"stablecode-completion-alpha-3b","name":"Stablecode Completion Alpha 3b","created_by":"stability","source":"official","description":"Stablecode Completion Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-31","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablecode-completion-alpha-3b","open_weight":true,"tagline":"Stablecode Completion Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":539,"last_modified":"2023-08-08","last_updated":"2026-06-20"},{"id":"stablecode-instruct-alpha-3b","name":"Stablecode Instruct Alpha 3b","created_by":"stability","source":"official","description":"Stablecode Instruct Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-07","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablecode-instruct-alpha-3b","open_weight":true,"tagline":"Stablecode Instruct Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":6,"last_modified":"2023-08-08","last_updated":"2026-06-18"},{"id":"stablelm-2-12b-chat-GGUF","name":"Stablelm 2 12b Chat GGUF","created_by":"stability","source":"official","description":"Stablelm 2 12b Chat GGUF is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-17","model_type":"chat","license":"other","parameters":12,"page_url":"https://huggingface.co/stabilityai/stablelm-2-12b-chat-GGUF","open_weight":true,"tagline":"Stablelm 2 12b Chat GGUF is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"huggingface_downloads":91,"last_modified":"2024-04-20","last_updated":"2026-06-20"},{"id":"stablelm-2-12b-chat","name":"Stablelm 2 12b Chat","created_by":"stability","source":"official","description":"Stablelm 2 12b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-04","model_type":"chat","license":"other","parameters":12,"page_url":"https://huggingface.co/stabilityai/stablelm-2-12b-chat","open_weight":true,"tagline":"Stablelm 2 12b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"huggingface_downloads":259,"last_modified":"2024-05-20","last_updated":"2026-06-20"},{"id":"stablelm-2-12b","name":"Stablelm 2 12b","created_by":"stability","source":"official","description":"Stablelm 2 12b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-21","model_type":"chat","license":"other","parameters":12,"page_url":"https://huggingface.co/stabilityai/stablelm-2-12b","open_weight":true,"tagline":"Stablelm 2 12b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1105,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stablelm-2-1_6b-chat","name":"Stablelm 2 1 6b Chat","created_by":"stability","source":"official","description":"Stablelm 2 1 6b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-08","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/stablelm-2-1_6b-chat","open_weight":true,"tagline":"Stablelm 2 1 6b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":910,"last_modified":"2024-06-03","last_updated":"2026-06-20"},{"id":"stablelm-2-1_6b","name":"Stablelm 2 1 6b","created_by":"stability","source":"official","description":"Stablelm 2 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-01-18","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/stablelm-2-1_6b","open_weight":true,"tagline":"Stablelm 2 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":4807,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"stablelm-2-zephyr-1_6b","name":"Stablelm 2 Zephyr 1 6b","created_by":"stability","source":"official","description":"Stablelm 2 Zephyr 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-01-19","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/stablelm-2-zephyr-1_6b","open_weight":true,"tagline":"Stablelm 2 Zephyr 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3143,"last_modified":"2024-06-03","last_updated":"2026-06-20"},{"id":"stablelm-3b-4e1t","name":"Stablelm 3b 4e1t","created_by":"stability","source":"official","description":"Stablelm 3b 4e1t is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-09-29","model_type":"chat","license":"cc-by-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-3b-4e1t","open_weight":true,"tagline":"Stablelm 3b 4e1t is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":96330,"last_modified":"2024-03-07","last_updated":"2026-06-20"},{"id":"stablelm-base-alpha-3b-v2","name":"Stablelm Base Alpha 3b V2","created_by":"stability","source":"official","description":"Stablelm Base Alpha 3b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-04","model_type":"chat","license":"cc-by-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-3b-v2","open_weight":true,"tagline":"Stablelm Base Alpha 3b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":528,"last_modified":"2023-09-11","last_updated":"2026-06-20"},{"id":"stablelm-base-alpha-3b","name":"Stablelm Base Alpha 3b","created_by":"stability","source":"official","description":"Stablelm Base Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-17","model_type":"chat","license":"cc-by-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-3b","open_weight":true,"tagline":"Stablelm Base Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":2523,"last_modified":"2023-10-19","last_updated":"2026-06-20"},{"id":"stablelm-base-alpha-7b-v2","name":"Stablelm Base Alpha 7b V2","created_by":"stability","source":"official","description":"Stablelm Base Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-04","model_type":"chat","license":"cc-by-sa-4.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-7b-v2","open_weight":true,"tagline":"Stablelm Base Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":532,"last_modified":"2023-09-11","last_updated":"2026-06-20"},{"id":"stablelm-base-alpha-7b","name":"Stablelm Base Alpha 7b","created_by":"stability","source":"official","description":"Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-11","model_type":"chat","license":"cc-by-sa-4.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-7b","open_weight":true,"tagline":"Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1231,"last_modified":"2023-10-19","last_updated":"2026-06-20"},{"id":"stablelm-tuned-alpha-3b","name":"Stablelm Tuned Alpha 3b","created_by":"stability","source":"official","description":"Stablelm Tuned Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-19","model_type":"chat","license":"cc-by-nc-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-tuned-alpha-3b","open_weight":true,"tagline":"Stablelm Tuned Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1002,"last_modified":"2023-04-19","last_updated":"2026-06-20"},{"id":"stablelm-tuned-alpha-7b","name":"Stablelm Tuned Alpha 7b","created_by":"stability","source":"official","description":"Stablelm Tuned Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-19","model_type":"chat","license":"cc-by-nc-sa-4.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/stablelm-tuned-alpha-7b","open_weight":true,"tagline":"Stablelm Tuned Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":5873,"last_modified":"2023-04-19","last_updated":"2026-06-20"},{"id":"stablelm-zephyr-3b","name":"Stablelm Zephyr 3b","created_by":"stability","source":"official","description":"Stablelm Zephyr 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-21","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-zephyr-3b","open_weight":true,"tagline":"Stablelm Zephyr 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":28832,"last_modified":"2024-07-10","last_updated":"2026-06-20"},{"id":"sv3d","name":"Sv3d","created_by":"stability","source":"official","description":"Sv3d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-18","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/sv3d","open_weight":true,"tagline":"Sv3d is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":0,"last_modified":"2024-07-10","last_updated":"2026-05-11"},{"id":"sv4d","name":"Sv4d","created_by":"stability","source":"official","description":"Sv4d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-07-16","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/sv4d","open_weight":true,"tagline":"Sv4d is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":24,"last_modified":"2024-08-03","last_updated":"2026-06-20"},{"id":"sv4d2.0","name":"Sv4d2.0","created_by":"stability","source":"official","description":"Sv4d2.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-02-20","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/sv4d2.0","open_weight":true,"tagline":"Sv4d2.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":436,"last_modified":"2025-04-04","last_updated":"2026-06-20"},{"id":"tiny-random-stablelm-2","name":"Tiny Random Stablelm 2","created_by":"stability","source":"official","description":"Tiny Random Stablelm 2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-20","model_type":"chat","page_url":"https://huggingface.co/stabilityai/tiny-random-stablelm-2","open_weight":true,"tagline":"Tiny Random Stablelm 2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":901,"last_modified":"2024-03-31","last_updated":"2026-06-20"},{"id":"TripoSR","name":"TripoSR","created_by":"stability","source":"official","description":"TripoSR is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-29","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/TripoSR","open_weight":true,"tagline":"TripoSR is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"huggingface_downloads":164680,"last_modified":"2024-08-09","last_updated":"2026-06-20"}]};
+export const provider: ProviderWithModels = {"id":"stability","name":"Stability AI","region":"GB","headquarters":"London, United Kingdom","founded":2019,"url":"https://stability.ai","api_url":"https://api.stability.ai","docs_url":"https://platform.stability.ai/docs","pricing_url":"https://platform.stability.ai/pricing","description":"AI company behind Stable Diffusion, Stable Video, Stable Audio, and related open and hosted generative models.","type":"direct","models_url":"https://huggingface.co/stabilityai","openai_compatible":false,"free_tier":false,"github_url":"https://github.com/Stability-AI","twitter_url":"https://x.com/StabilityAI","icon":"","models":[{"id":"ar-stablelm-2-base","name":"Ar Stablelm 2 Base","created_by":"stability","source":"official","description":"Ar Stablelm 2 Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-07-29","model_type":"chat","license":"other","page_url":"https://huggingface.co/stabilityai/ar-stablelm-2-base","open_weight":true,"tagline":"Ar Stablelm 2 Base is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":2,"last_modified":"2024-12-06","last_updated":"2026-06-03"},{"id":"ar-stablelm-2-chat-gguf","name":"Ar Stablelm 2 Chat Gguf","created_by":"stability","source":"official","description":"Ar Stablelm 2 Chat Gguf is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-08-09","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/ar-stablelm-2-chat-gguf","open_weight":true,"tagline":"Ar Stablelm 2 Chat Gguf is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":6,"last_modified":"2024-12-20","last_updated":"2026-06-03"},{"id":"ar-stablelm-2-chat","name":"Ar Stablelm 2 Chat","created_by":"stability","source":"official","description":"Ar Stablelm 2 Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-08-06","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/ar-stablelm-2-chat","open_weight":true,"tagline":"Ar Stablelm 2 Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":7,"last_modified":"2024-12-06","last_updated":"2026-06-29"},{"id":"arcade100k","name":"Arcade100k","created_by":"stability","source":"official","description":"Arcade100k is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-27","model_type":"other","page_url":"https://huggingface.co/stabilityai/arcade100k","open_weight":true,"tagline":"Arcade100k is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":0,"last_modified":"2024-04-12","last_updated":"2026-05-11"},{"id":"codellama13b_instruct_260k_synthesis","name":"Codellama13b Instruct 260k Synthesis","created_by":"stability","source":"official","family":"codellama","description":"Codellama13b Instruct 260k Synthesis is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-08","model_type":"chat","page_url":"https://huggingface.co/stabilityai/codellama13b_instruct_260k_synthesis","open_weight":true,"tagline":"Codellama13b Instruct 260k Synthesis is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":38,"last_modified":"2023-11-08","last_updated":"2026-06-20"},{"id":"control-lora","name":"Control Lora","created_by":"stability","source":"official","description":"Control Lora is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-17","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/control-lora","open_weight":true,"tagline":"Control Lora is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2023-08-19","last_updated":"2026-05-11"},{"id":"cosxl","name":"Cosxl","created_by":"stability","source":"official","description":"Cosxl is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-08","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/cosxl","open_weight":true,"tagline":"Cosxl is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2024-04-13","last_updated":"2026-05-11"},{"id":"japanese-instructblip-alpha","name":"Japanese Instructblip Alpha","created_by":"stability","source":"official","description":"Japanese Instructblip Alpha is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-15","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-instructblip-alpha","open_weight":true,"tagline":"Japanese Instructblip Alpha is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":59,"last_modified":"2023-11-17","last_updated":"2026-06-29"},{"id":"japanese-stable-clip-vit-l-16","name":"Japanese Stable Clip Vit L 16","created_by":"stability","source":"official","description":"Japanese Stable Clip Vit L 16 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-01","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-stable-clip-vit-l-16","open_weight":true,"tagline":"Japanese Stable Clip Vit L 16 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1660,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"japanese-stable-diffusion-xl","name":"Japanese Stable Diffusion Xl","created_by":"stability","source":"official","description":"Japanese Stable Diffusion Xl is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-01","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-stable-diffusion-xl","open_weight":true,"tagline":"Japanese Stable Diffusion Xl is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":45,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"japanese-stable-vlm","name":"Japanese Stable Vlm","created_by":"stability","source":"official","description":"Japanese Stable Vlm is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-01","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/japanese-stable-vlm","open_weight":true,"tagline":"Japanese Stable Vlm is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":13,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"japanese-stablelm-2-base-1_6b","name":"Japanese Stablelm 2 Base 1 6b","created_by":"stability","source":"official","description":"Japanese Stablelm 2 Base 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-02","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-2-base-1_6b","open_weight":true,"tagline":"Japanese Stablelm 2 Base 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3,"last_modified":"2024-05-02","last_updated":"2026-06-03"},{"id":"japanese-stablelm-2-instruct-1_6b","name":"Japanese Stablelm 2 Instruct 1 6b","created_by":"stability","source":"official","description":"Japanese Stablelm 2 Instruct 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-02","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-2-instruct-1_6b","open_weight":true,"tagline":"Japanese Stablelm 2 Instruct 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":68,"last_modified":"2024-07-10","last_updated":"2026-06-03"},{"id":"japanese-stablelm-3b-4e1t-base","name":"Japanese Stablelm 3b 4e1t Base","created_by":"stability","source":"official","description":"Japanese Stablelm 3b 4e1t Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-base","open_weight":true,"tagline":"Japanese Stablelm 3b 4e1t Base is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":48,"last_modified":"2024-04-26","last_updated":"2026-06-29"},{"id":"japanese-stablelm-3b-4e1t-instruct","name":"Japanese Stablelm 3b 4e1t Instruct","created_by":"stability","source":"official","description":"Japanese Stablelm 3b 4e1t Instruct is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-3b-4e1t-instruct","open_weight":true,"tagline":"Japanese Stablelm 3b 4e1t Instruct is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":148,"last_modified":"2024-04-26","last_updated":"2026-06-29"},{"id":"japanese-stablelm-base-alpha-7b","name":"Japanese Stablelm Base Alpha 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-09","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-alpha-7b","open_weight":true,"tagline":"Japanese Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":330,"last_modified":"2023-08-22","last_updated":"2026-06-29"},{"id":"japanese-stablelm-base-beta-70b","name":"Japanese Stablelm Base Beta 70b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":70,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-beta-70b","open_weight":true,"tagline":"Japanese Stablelm Base Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":127,"last_modified":"2023-12-19","last_updated":"2026-06-29"},{"id":"japanese-stablelm-base-beta-7b","name":"Japanese Stablelm Base Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Base Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":50,"last_modified":"2023-12-19","last_updated":"2026-06-29"},{"id":"japanese-stablelm-base-gamma-7b","name":"Japanese Stablelm Base Gamma 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-gamma-7b","open_weight":true,"tagline":"Japanese Stablelm Base Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":141,"last_modified":"2024-01-25","last_updated":"2026-06-29"},{"id":"japanese-stablelm-base-ja_vocab-beta-7b","name":"Japanese Stablelm Base Ja Vocab Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Base Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-base-ja_vocab-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Base Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":40,"last_modified":"2023-12-19","last_updated":"2026-06-29"},{"id":"japanese-stablelm-instruct-alpha-7b-v2","name":"Japanese Stablelm Instruct Alpha 7b V2","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-09-26","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b-v2","open_weight":true,"tagline":"Japanese Stablelm Instruct Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":40,"last_modified":"2023-10-06","last_updated":"2026-06-29"},{"id":"japanese-stablelm-instruct-alpha-7b","name":"Japanese Stablelm Instruct Alpha 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-10","model_type":"chat","license":"other","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-alpha-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":0,"last_modified":"2023-08-22","last_updated":"2026-05-11"},{"id":"japanese-stablelm-instruct-beta-70b","name":"Japanese Stablelm Instruct Beta 70b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":70,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-70b","open_weight":true,"tagline":"Japanese Stablelm Instruct Beta 70b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":131,"last_modified":"2023-12-19","last_updated":"2026-06-29"},{"id":"japanese-stablelm-instruct-beta-7b","name":"Japanese Stablelm Instruct Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":58,"last_modified":"2023-12-19","last_updated":"2026-06-29"},{"id":"japanese-stablelm-instruct-gamma-7b","name":"Japanese Stablelm Instruct Gamma 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-16","model_type":"chat","license":"apache-2.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-gamma-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Gamma 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":151,"last_modified":"2024-01-24","last_updated":"2026-06-29"},{"id":"japanese-stablelm-instruct-ja_vocab-beta-7b","name":"Japanese Stablelm Instruct Ja Vocab Beta 7b","created_by":"stability","source":"official","description":"Japanese Stablelm Instruct Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-10-30","model_type":"chat","license":"llama2","parameters":7,"page_url":"https://huggingface.co/stabilityai/japanese-stablelm-instruct-ja_vocab-beta-7b","open_weight":true,"tagline":"Japanese Stablelm Instruct Ja Vocab Beta 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":45,"last_modified":"2023-12-19","last_updated":"2026-06-29"},{"id":"SAME-L","name":"SAME L","created_by":"stability","source":"official","description":"SAME L is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/SAME-L","open_weight":true,"tagline":"SAME L is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":6323,"last_modified":"2026-06-24","last_updated":"2026-06-29"},{"id":"SAME-S","name":"SAME S","created_by":"stability","source":"official","description":"SAME S is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"code","license":"other","page_url":"https://huggingface.co/stabilityai/SAME-S","open_weight":true,"tagline":"SAME S is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1549,"last_modified":"2026-05-19","last_updated":"2026-06-29"},{"id":"sd-turbo","name":"Sd Turbo","created_by":"stability","source":"official","description":"Sd Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-27","model_type":"image","page_url":"https://huggingface.co/stabilityai/sd-turbo","open_weight":true,"tagline":"Sd Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":698263,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"sd-vae-ft-ema-original","name":"Sd Vae Ft Ema Original","created_by":"stability","source":"official","description":"Sd Vae Ft Ema Original is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"image","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-ema-original","open_weight":true,"tagline":"Sd Vae Ft Ema Original is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2022-12-19","last_updated":"2026-05-11"},{"id":"sd-vae-ft-ema","name":"Sd Vae Ft Ema","created_by":"stability","source":"official","description":"Sd Vae Ft Ema is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-ema","open_weight":true,"tagline":"Sd Vae Ft Ema is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":7037,"last_modified":"2023-06-05","last_updated":"2026-06-29"},{"id":"sd-vae-ft-mse-original","name":"Sd Vae Ft Mse Original","created_by":"stability","source":"official","description":"Sd Vae Ft Mse Original is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"image","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-mse-original","open_weight":true,"tagline":"Sd Vae Ft Mse Original is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":1,"last_modified":"2022-12-19","last_updated":"2026-06-29"},{"id":"sd-vae-ft-mse","name":"Sd Vae Ft Mse","created_by":"stability","source":"official","description":"Sd Vae Ft Mse is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-10-13","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/sd-vae-ft-mse","open_weight":true,"tagline":"Sd Vae Ft Mse is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":102468,"last_modified":"2023-06-06","last_updated":"2026-06-29"},{"id":"sd-x2-latent-upscaler","name":"Sd X2 Latent Upscaler","created_by":"stability","source":"official","description":"Sd X2 Latent Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-02-03","model_type":"other","license":"openrail++","page_url":"https://huggingface.co/stabilityai/sd-x2-latent-upscaler","open_weight":true,"tagline":"Sd X2 Latent Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":130271,"last_modified":"2023-06-05","last_updated":"2026-06-29"},{"id":"sdxl-turbo-ryzen-ai","name":"Sdxl Turbo Ryzen Ai","created_by":"stability","source":"official","description":"Sdxl Turbo Ryzen Ai is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-12-06","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/sdxl-turbo-ryzen-ai","open_weight":true,"tagline":"Sdxl Turbo Ryzen Ai is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":489,"last_modified":"2024-12-11","last_updated":"2026-06-29"},{"id":"sdxl-turbo-tensorrt","name":"Sdxl Turbo Tensorrt","created_by":"stability","source":"official","description":"Sdxl Turbo Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-12-28","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/sdxl-turbo-tensorrt","open_weight":true,"tagline":"Sdxl Turbo Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":29,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"sdxl-turbo","name":"Sdxl Turbo","created_by":"stability","source":"official","description":"Sdxl Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-27","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/sdxl-turbo","open_weight":true,"tagline":"Sdxl Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":609582,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"sdxl-vae","name":"Sdxl Vae","created_by":"stability","source":"official","description":"Sdxl Vae is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-06-21","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/sdxl-vae","open_weight":true,"tagline":"Sdxl Vae is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":263433,"last_modified":"2023-08-04","last_updated":"2026-06-29"},{"id":"sp4d","name":"Sp4d","created_by":"stability","source":"official","description":"Sp4d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-11-03","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/sp4d","open_weight":true,"tagline":"Sp4d is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":15,"last_modified":"2025-11-05","last_updated":"2026-06-29"},{"id":"stable-audio-3-medium-base","name":"Stable Audio 3 Medium Base","created_by":"stability","source":"official","description":"Stable Audio 3 Medium Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-medium-base","open_weight":true,"tagline":"Stable Audio 3 Medium Base is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":4312,"last_modified":"2026-06-16","last_updated":"2026-06-29"},{"id":"stable-audio-3-medium","name":"Stable Audio 3 Medium","created_by":"stability","source":"official","description":"Stable Audio 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-medium","open_weight":true,"tagline":"Stable Audio 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":59946,"last_modified":"2026-06-16","last_updated":"2026-06-29"},{"id":"stable-audio-3-small-music-base","name":"Stable Audio 3 Small Music Base","created_by":"stability","source":"official","description":"Stable Audio 3 Small Music Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-16","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-music-base","open_weight":true,"tagline":"Stable Audio 3 Small Music Base is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":1769,"last_modified":"2026-05-20","last_updated":"2026-06-29"},{"id":"stable-audio-3-small-music","name":"Stable Audio 3 Small Music","created_by":"stability","source":"official","description":"Stable Audio 3 Small Music is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-music","open_weight":true,"tagline":"Stable Audio 3 Small Music is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":15603,"last_modified":"2026-05-19","last_updated":"2026-06-29"},{"id":"stable-audio-3-small-sfx-base","name":"Stable Audio 3 Small Sfx Base","created_by":"stability","source":"official","description":"Stable Audio 3 Small Sfx Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-sfx-base","open_weight":true,"tagline":"Stable Audio 3 Small Sfx Base is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":1054,"last_modified":"2026-05-20","last_updated":"2026-06-29"},{"id":"stable-audio-3-small-sfx","name":"Stable Audio 3 Small Sfx","created_by":"stability","source":"official","description":"Stable Audio 3 Small Sfx is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-17","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-3-small-sfx","open_weight":true,"tagline":"Stable Audio 3 Small Sfx is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":12006,"last_modified":"2026-05-19","last_updated":"2026-06-29"},{"id":"stable-audio-open-1.0","name":"Stable Audio Open 1.0","created_by":"stability","source":"official","description":"Stable Audio Open 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-24","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-open-1.0","open_weight":true,"tagline":"Stable Audio Open 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":26532,"last_modified":"2025-06-19","last_updated":"2026-06-29"},{"id":"stable-audio-open-small","name":"Stable Audio Open Small","created_by":"stability","source":"official","description":"Stable Audio Open Small is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-05-12","model_type":"audio","license":"other","page_url":"https://huggingface.co/stabilityai/stable-audio-open-small","open_weight":true,"tagline":"Stable Audio Open Small is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":2235,"last_modified":"2025-05-27","last_updated":"2026-06-29"},{"id":"stable-cascade-prior","name":"Stable Cascade Prior","created_by":"stability","source":"official","description":"Stable Cascade Prior is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-12","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-cascade-prior","open_weight":true,"tagline":"Stable Cascade Prior is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":789,"last_modified":"2024-03-14","last_updated":"2026-06-29"},{"id":"stable-cascade","name":"Stable Cascade","created_by":"stability","source":"official","description":"Stable Cascade is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-06","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-cascade","open_weight":true,"tagline":"Stable Cascade is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":8252,"last_modified":"2024-03-16","last_updated":"2026-06-29"},{"id":"stable-code-3b","name":"Stable Code 3b","created_by":"stability","source":"official","description":"Stable Code 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-01-09","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stable-code-3b","open_weight":true,"tagline":"Stable Code 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":5918,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stable-code-instruct-3b","name":"Stable Code Instruct 3b","created_by":"stability","source":"official","description":"Stable Code Instruct 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-06","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stable-code-instruct-3b","open_weight":true,"tagline":"Stable Code Instruct 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":2335,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stable-codec-speech-16k-base","name":"Stable Codec Speech 16k Base","created_by":"stability","source":"official","description":"Stable Codec Speech 16k Base is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-01-14","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-codec-speech-16k-base","open_weight":true,"tagline":"Stable Codec Speech 16k Base is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":1,"last_modified":"2025-01-14","last_updated":"2026-06-03"},{"id":"stable-codec-speech-16k","name":"Stable Codec Speech 16k","created_by":"stability","source":"official","description":"Stable Codec Speech 16k is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-01-10","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-codec-speech-16k","open_weight":true,"tagline":"Stable Codec Speech 16k is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"fine_tuning":true},"tools":["function_calling"],"huggingface_downloads":334,"last_modified":"2025-01-10","last_updated":"2026-06-29"},{"id":"stable-diffusion-3-medium-amdnpu","name":"Stable Diffusion 3 Medium Amdnpu","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Amdnpu is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2026-05-01","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium-amdnpu","open_weight":true,"tagline":"Stable Diffusion 3 Medium Amdnpu is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":27,"last_modified":"2026-05-05","last_updated":"2026-06-29"},{"id":"stable-diffusion-3-medium-diffusers","name":"Stable Diffusion 3 Medium Diffusers","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Diffusers is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-06-12","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium-diffusers","open_weight":true,"tagline":"Stable Diffusion 3 Medium Diffusers is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":33289,"last_modified":"2024-06-19","last_updated":"2026-06-29"},{"id":"stable-diffusion-3-medium-tensorrt","name":"Stable Diffusion 3 Medium Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-06-12","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":15,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"stable-diffusion-3-medium","name":"Stable Diffusion 3 Medium","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-05-30","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium","open_weight":true,"tagline":"Stable Diffusion 3 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":3481,"last_modified":"2024-08-12","last_updated":"2026-06-29"},{"id":"stable-diffusion-3-medium_amdgpu","name":"Stable Diffusion 3 Medium Amdgpu","created_by":"stability","source":"official","description":"Stable Diffusion 3 Medium Amdgpu is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-02-28","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3-medium_amdgpu","open_weight":true,"tagline":"Stable Diffusion 3 Medium Amdgpu is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":0,"last_modified":"2025-07-16","last_updated":"2026-05-11"},{"id":"stable-diffusion-3.5-controlnets-tensorrt","name":"Stable Diffusion 3.5 Controlnets Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Controlnets Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-08-07","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-controlnets-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3.5 Controlnets Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":177,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-controlnets","name":"Stable Diffusion 3.5 Controlnets","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Controlnets is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-15","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-controlnets","open_weight":true,"tagline":"Stable Diffusion 3.5 Controlnets is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":12,"last_modified":"2024-12-11","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-large-controlnet-blur","name":"Stable Diffusion 3.5 Large Controlnet Blur","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Controlnet Blur is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-25","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-controlnet-blur","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Controlnet Blur is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":53,"last_modified":"2024-11-28","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-large-controlnet-canny","name":"Stable Diffusion 3.5 Large Controlnet Canny","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Controlnet Canny is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-25","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-controlnet-canny","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Controlnet Canny is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":21152,"last_modified":"2024-11-28","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-large-controlnet-depth","name":"Stable Diffusion 3.5 Large Controlnet Depth","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Controlnet Depth is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-11-25","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-controlnet-depth","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Controlnet Depth is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":24549,"last_modified":"2024-11-28","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-large-tensorrt","name":"Stable Diffusion 3.5 Large Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-05-16","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":1469,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-large-turbo","name":"Stable Diffusion 3.5 Large Turbo","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-10-22","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large-turbo","open_weight":true,"tagline":"Stable Diffusion 3.5 Large Turbo is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":6991,"last_modified":"2024-10-22","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-large","name":"Stable Diffusion 3.5 Large","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Large is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-10-22","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-large","open_weight":true,"tagline":"Stable Diffusion 3.5 Large is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":26298,"last_modified":"2024-10-22","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-medium-tensorrt","name":"Stable Diffusion 3.5 Medium Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-08-07","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-medium-tensorrt","open_weight":true,"tagline":"Stable Diffusion 3.5 Medium Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":9,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"stable-diffusion-3.5-medium","name":"Stable Diffusion 3.5 Medium","created_by":"stability","source":"official","description":"Stable Diffusion 3.5 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-10-29","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-3.5-medium","open_weight":true,"tagline":"Stable Diffusion 3.5 Medium is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":612474,"last_modified":"2024-10-31","last_updated":"2026-06-29"},{"id":"stable-diffusion-x4-upscaler","name":"Stable Diffusion X4 Upscaler","created_by":"stability","source":"official","description":"Stable Diffusion X4 Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2022-11-23","model_type":"other","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-x4-upscaler","open_weight":true,"tagline":"Stable Diffusion X4 Upscaler is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":10328,"last_modified":"2023-07-05","last_updated":"2026-06-29"},{"id":"stable-diffusion-xl-1.0-tensorrt","name":"Stable Diffusion Xl 1.0 Tensorrt","created_by":"stability","source":"official","description":"Stable Diffusion Xl 1.0 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-22","model_type":"image","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-1.0-tensorrt","open_weight":true,"tagline":"Stable Diffusion Xl 1.0 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":275,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"stable-diffusion-xl-base-0.9","name":"Stable Diffusion Xl Base 0.9","created_by":"stability","source":"official","description":"Stable Diffusion Xl Base 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-06-19","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-base-0.9","open_weight":true,"tagline":"Stable Diffusion Xl Base 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":48,"last_modified":"2023-07-12","last_updated":"2026-06-29"},{"id":"stable-diffusion-xl-base-1.0","name":"Stable Diffusion Xl Base 1.0","created_by":"stability","source":"official","description":"Stable Diffusion Xl Base 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-25","model_type":"image","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0","open_weight":true,"tagline":"Stable Diffusion Xl Base 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":1324082,"last_modified":"2023-10-30","last_updated":"2026-06-29"},{"id":"stable-diffusion-xl-refiner-0.9","name":"Stable Diffusion Xl Refiner 0.9","created_by":"stability","source":"official","description":"Stable Diffusion Xl Refiner 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-06-21","model_type":"image","license":"other","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-0.9","open_weight":true,"tagline":"Stable Diffusion Xl Refiner 0.9 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":45,"last_modified":"2023-07-12","last_updated":"2026-06-29"},{"id":"stable-diffusion-xl-refiner-1.0","name":"Stable Diffusion Xl Refiner 1.0","created_by":"stability","source":"official","description":"Stable Diffusion Xl Refiner 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-26","model_type":"image","license":"openrail++","page_url":"https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0","open_weight":true,"tagline":"Stable Diffusion Xl Refiner 1.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["image"]},"endpoints":["images"],"huggingface_downloads":163793,"last_modified":"2023-09-25","last_updated":"2026-06-29"},{"id":"stable-fast-3d","name":"Stable Fast 3d","created_by":"stability","source":"official","description":"Stable Fast 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-07-17","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-fast-3d","open_weight":true,"tagline":"Stable Fast 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"huggingface_downloads":11960,"last_modified":"2025-04-08","last_updated":"2026-06-29"},{"id":"stable-point-aware-3d","name":"Stable Point Aware 3d","created_by":"stability","source":"official","description":"Stable Point Aware 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-12-28","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-point-aware-3d","open_weight":true,"tagline":"Stable Point Aware 3d is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"huggingface_downloads":2132,"last_modified":"2025-04-08","last_updated":"2026-06-29"},{"id":"stable-video-diffusion-img2vid-xt-1-1-tensorrt","name":"Stable Video Diffusion Img2vid Xt 1 1 Tensorrt","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid Xt 1 1 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-07","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt-1-1-tensorrt","open_weight":true,"tagline":"Stable Video Diffusion Img2vid Xt 1 1 Tensorrt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":20,"last_modified":"2025-10-20","last_updated":"2026-06-29"},{"id":"stable-video-diffusion-img2vid-xt-1-1","name":"Stable Video Diffusion Img2vid Xt 1 1","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid Xt 1 1 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-02","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt-1-1","open_weight":true,"tagline":"Stable Video Diffusion Img2vid Xt 1 1 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":7772,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stable-video-diffusion-img2vid-xt","name":"Stable Video Diffusion Img2vid Xt","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid Xt is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-20","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt","open_weight":true,"tagline":"Stable Video Diffusion Img2vid Xt is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":218304,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stable-video-diffusion-img2vid","name":"Stable Video Diffusion Img2vid","created_by":"stability","source":"official","description":"Stable Video Diffusion Img2vid is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-20","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-video-diffusion-img2vid","open_weight":true,"tagline":"Stable Video Diffusion Img2vid is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":32007,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stable-virtual-camera","name":"Stable Virtual Camera","created_by":"stability","source":"official","description":"Stable Virtual Camera is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-02-19","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/stable-virtual-camera","open_weight":true,"tagline":"Stable Virtual Camera is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":9634,"last_modified":"2025-06-03","last_updated":"2026-06-29"},{"id":"stable-zero123","name":"Stable Zero123","created_by":"stability","source":"official","description":"Stable Zero123 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-12-08","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/stable-zero123","open_weight":true,"tagline":"Stable Zero123 is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":0,"last_modified":"2024-07-10","last_updated":"2026-05-11"},{"id":"StableBeluga-13B","name":"StableBeluga 13B","created_by":"stability","source":"official","description":"StableBeluga 13B is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-27","model_type":"chat","parameters":13,"page_url":"https://huggingface.co/stabilityai/StableBeluga-13B","open_weight":true,"tagline":"StableBeluga 13B is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":260,"last_modified":"2023-08-29","last_updated":"2026-06-29"},{"id":"StableBeluga-7B","name":"StableBeluga 7B","created_by":"stability","source":"official","description":"StableBeluga 7B is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-27","model_type":"chat","parameters":7,"page_url":"https://huggingface.co/stabilityai/StableBeluga-7B","open_weight":true,"tagline":"StableBeluga 7B is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":251,"last_modified":"2023-08-29","last_updated":"2026-06-29"},{"id":"StableBeluga1-Delta","name":"StableBeluga1 Delta","created_by":"stability","source":"official","description":"StableBeluga1 Delta is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-20","model_type":"chat","license":"cc-by-nc-4.0","page_url":"https://huggingface.co/stabilityai/StableBeluga1-Delta","open_weight":true,"tagline":"StableBeluga1 Delta is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":145,"last_modified":"2023-07-27","last_updated":"2026-06-29"},{"id":"StableBeluga2","name":"StableBeluga2","created_by":"stability","source":"official","description":"StableBeluga2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-20","model_type":"chat","page_url":"https://huggingface.co/stabilityai/StableBeluga2","open_weight":true,"tagline":"StableBeluga2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":717,"last_modified":"2023-09-18","last_updated":"2026-06-29"},{"id":"stablecode-completion-alpha-3b-4k","name":"Stablecode Completion Alpha 3b 4k","created_by":"stability","source":"official","description":"Stablecode Completion Alpha 3b 4k is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-07","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablecode-completion-alpha-3b-4k","open_weight":true,"tagline":"Stablecode Completion Alpha 3b 4k is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1115,"last_modified":"2023-08-08","last_updated":"2026-06-29"},{"id":"stablecode-completion-alpha-3b","name":"Stablecode Completion Alpha 3b","created_by":"stability","source":"official","description":"Stablecode Completion Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-07-31","model_type":"chat","license":"apache-2.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablecode-completion-alpha-3b","open_weight":true,"tagline":"Stablecode Completion Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":408,"last_modified":"2023-08-08","last_updated":"2026-06-29"},{"id":"stablecode-instruct-alpha-3b","name":"Stablecode Instruct Alpha 3b","created_by":"stability","source":"official","description":"Stablecode Instruct Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-07","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablecode-instruct-alpha-3b","open_weight":true,"tagline":"Stablecode Instruct Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":4,"last_modified":"2023-08-08","last_updated":"2026-06-29"},{"id":"stablelm-2-12b-chat-GGUF","name":"Stablelm 2 12b Chat GGUF","created_by":"stability","source":"official","description":"Stablelm 2 12b Chat GGUF is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-17","model_type":"chat","license":"other","parameters":12,"page_url":"https://huggingface.co/stabilityai/stablelm-2-12b-chat-GGUF","open_weight":true,"tagline":"Stablelm 2 12b Chat GGUF is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"huggingface_downloads":168,"last_modified":"2024-04-20","last_updated":"2026-06-29"},{"id":"stablelm-2-12b-chat","name":"Stablelm 2 12b Chat","created_by":"stability","source":"official","description":"Stablelm 2 12b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-04","model_type":"chat","license":"other","parameters":12,"page_url":"https://huggingface.co/stabilityai/stablelm-2-12b-chat","open_weight":true,"tagline":"Stablelm 2 12b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"tool_call":true,"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"tools":["function_calling"],"huggingface_downloads":266,"last_modified":"2024-05-20","last_updated":"2026-06-29"},{"id":"stablelm-2-12b","name":"Stablelm 2 12b","created_by":"stability","source":"official","description":"Stablelm 2 12b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-21","model_type":"chat","license":"other","parameters":12,"page_url":"https://huggingface.co/stabilityai/stablelm-2-12b","open_weight":true,"tagline":"Stablelm 2 12b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3144,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stablelm-2-1_6b-chat","name":"Stablelm 2 1 6b Chat","created_by":"stability","source":"official","description":"Stablelm 2 1 6b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-04-08","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/stablelm-2-1_6b-chat","open_weight":true,"tagline":"Stablelm 2 1 6b Chat is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":843,"last_modified":"2024-06-03","last_updated":"2026-06-29"},{"id":"stablelm-2-1_6b","name":"Stablelm 2 1 6b","created_by":"stability","source":"official","description":"Stablelm 2 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-01-18","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/stablelm-2-1_6b","open_weight":true,"tagline":"Stablelm 2 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":10285,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"stablelm-2-zephyr-1_6b","name":"Stablelm 2 Zephyr 1 6b","created_by":"stability","source":"official","description":"Stablelm 2 Zephyr 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-01-19","model_type":"chat","license":"other","parameters":6,"page_url":"https://huggingface.co/stabilityai/stablelm-2-zephyr-1_6b","open_weight":true,"tagline":"Stablelm 2 Zephyr 1 6b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3318,"last_modified":"2024-06-03","last_updated":"2026-06-29"},{"id":"stablelm-3b-4e1t","name":"Stablelm 3b 4e1t","created_by":"stability","source":"official","description":"Stablelm 3b 4e1t is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-09-29","model_type":"chat","license":"cc-by-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-3b-4e1t","open_weight":true,"tagline":"Stablelm 3b 4e1t is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":93520,"last_modified":"2024-03-07","last_updated":"2026-06-29"},{"id":"stablelm-base-alpha-3b-v2","name":"Stablelm Base Alpha 3b V2","created_by":"stability","source":"official","description":"Stablelm Base Alpha 3b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-04","model_type":"chat","license":"cc-by-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-3b-v2","open_weight":true,"tagline":"Stablelm Base Alpha 3b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":531,"last_modified":"2023-09-11","last_updated":"2026-06-29"},{"id":"stablelm-base-alpha-3b","name":"Stablelm Base Alpha 3b","created_by":"stability","source":"official","description":"Stablelm Base Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-17","model_type":"chat","license":"cc-by-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-3b","open_weight":true,"tagline":"Stablelm Base Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":3011,"last_modified":"2023-10-19","last_updated":"2026-06-29"},{"id":"stablelm-base-alpha-7b-v2","name":"Stablelm Base Alpha 7b V2","created_by":"stability","source":"official","description":"Stablelm Base Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-08-04","model_type":"chat","license":"cc-by-sa-4.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-7b-v2","open_weight":true,"tagline":"Stablelm Base Alpha 7b V2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":457,"last_modified":"2023-09-11","last_updated":"2026-06-29"},{"id":"stablelm-base-alpha-7b","name":"Stablelm Base Alpha 7b","created_by":"stability","source":"official","description":"Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-11","model_type":"chat","license":"cc-by-sa-4.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/stablelm-base-alpha-7b","open_weight":true,"tagline":"Stablelm Base Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":999,"last_modified":"2023-10-19","last_updated":"2026-06-29"},{"id":"stablelm-tuned-alpha-3b","name":"Stablelm Tuned Alpha 3b","created_by":"stability","source":"official","description":"Stablelm Tuned Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-19","model_type":"chat","license":"cc-by-nc-sa-4.0","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-tuned-alpha-3b","open_weight":true,"tagline":"Stablelm Tuned Alpha 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":845,"last_modified":"2023-04-19","last_updated":"2026-06-29"},{"id":"stablelm-tuned-alpha-7b","name":"Stablelm Tuned Alpha 7b","created_by":"stability","source":"official","description":"Stablelm Tuned Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-04-19","model_type":"chat","license":"cc-by-nc-sa-4.0","parameters":7,"page_url":"https://huggingface.co/stabilityai/stablelm-tuned-alpha-7b","open_weight":true,"tagline":"Stablelm Tuned Alpha 7b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":5838,"last_modified":"2023-04-19","last_updated":"2026-06-29"},{"id":"stablelm-zephyr-3b","name":"Stablelm Zephyr 3b","created_by":"stability","source":"official","description":"Stablelm Zephyr 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2023-11-21","model_type":"chat","license":"other","parameters":3,"page_url":"https://huggingface.co/stabilityai/stablelm-zephyr-3b","open_weight":true,"tagline":"Stablelm Zephyr 3b is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true,"fine_tuning":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":28453,"last_modified":"2024-07-10","last_updated":"2026-06-29"},{"id":"sv3d","name":"Sv3d","created_by":"stability","source":"official","description":"Sv3d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-18","model_type":"video","license":"other","page_url":"https://huggingface.co/stabilityai/sv3d","open_weight":true,"tagline":"Sv3d is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["video"],"huggingface_downloads":0,"last_modified":"2024-07-10","last_updated":"2026-05-11"},{"id":"sv4d","name":"Sv4d","created_by":"stability","source":"official","description":"Sv4d is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-07-16","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/sv4d","open_weight":true,"tagline":"Sv4d is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":27,"last_modified":"2024-08-03","last_updated":"2026-06-29"},{"id":"sv4d2.0","name":"Sv4d2.0","created_by":"stability","source":"official","description":"Sv4d2.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2025-02-20","model_type":"other","license":"other","page_url":"https://huggingface.co/stabilityai/sv4d2.0","open_weight":true,"tagline":"Sv4d2.0 is a Stability AI model published on the official Stability AI Hugging Face organization.","huggingface_downloads":365,"last_modified":"2025-04-04","last_updated":"2026-06-29"},{"id":"tiny-random-stablelm-2","name":"Tiny Random Stablelm 2","created_by":"stability","source":"official","description":"Tiny Random Stablelm 2 is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-03-20","model_type":"chat","page_url":"https://huggingface.co/stabilityai/tiny-random-stablelm-2","open_weight":true,"tagline":"Tiny Random Stablelm 2 is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["chat_completions"],"huggingface_downloads":1405,"last_modified":"2024-03-31","last_updated":"2026-06-29"},{"id":"TripoSR","name":"TripoSR","created_by":"stability","source":"official","description":"TripoSR is a Stability AI model published on the official Stability AI Hugging Face organization.","status":"active","release_date":"2024-02-29","model_type":"other","license":"mit","page_url":"https://huggingface.co/stabilityai/TripoSR","open_weight":true,"tagline":"TripoSR is a Stability AI model published on the official Stability AI Hugging Face organization.","capabilities":{"vision":true,"fine_tuning":true},"huggingface_downloads":189582,"last_modified":"2024-08-09","last_updated":"2026-06-29"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/together.ts b/packages/npm/src/providers/together.ts
index 43a07c0a8..7e9c40800 100644
--- a/packages/npm/src/providers/together.ts
+++ b/packages/npm/src/providers/together.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"together","name":"Together AI","region":"US","headquarters":"San Francisco, CA","founded":2022,"url":"https://www.together.ai","api_url":"https://api.together.xyz/v1","docs_url":"https://docs.together.ai","pricing_url":"https://www.together.ai/pricing","description":"Open-source AI inference and fine-tuning platform with 100+ models.","type":"cloud","playground_url":"https://api.together.ai/playground","sdk":{"python":"together","javascript":"together-ai"},"openai_compatible":true,"free_tier":true,"github_url":"https://github.com/togethercomputer","models_url":"https://docs.together.ai/docs/serverless-models","twitter_url":"https://x.com/togethercompute","blog_url":"https://www.together.ai/blog","terms_url":"https://www.together.ai/terms-of-service","support_url":"https://www.together.ai/support","icon":"","models":[{"id":"s/cartesia-sonic-3","name":"Cartesia Sonic 3","created_by":"unknown","source":"official","page_url":"https://www.together.ai/models/cartesia-sonic-3","capabilities":{"streaming":true},"last_updated":"2026-03-27"},{"id":"s/deepseek-v3-1","name":"Deepseek V3 1","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.6,"output":1.7},"page_url":"https://www.together.ai/models/deepseek-v3-1","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"s/deepseek-v3-2-exp","name":"Deepseek V3 2 Exp","created_by":"deepseek","source":"official","last_updated":"2026-04-10","family":"deepseek","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"page_url":"https://www.together.ai/models/deepseek-v3-2-exp","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"s/deepseek-v4-flash","name":"Deepseek V4 Flash","created_by":"deepseek","source":"official","family":"deepseek","page_url":"https://www.together.ai/models/deepseek-v4-flash","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.3,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03","description":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","context_window":1000000,"max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"s/deepseek-v4-pro","name":"Deepseek V4 Pro","created_by":"deepseek","source":"official","family":"deepseek","context_window":512000,"page_url":"https://www.together.ai/models/deepseek-v4-pro","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.2,"output":3.48},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"s/flux-2-pro","name":"Flux 2 Pro","created_by":"black-forest-labs","source":"official","last_updated":"2026-04-12","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.134,"output":0.03},"model_type":"image","page_url":"https://www.together.ai/models/flux-2-pro","description":"Production-grade default model for FLUX.2 text-to-image generation and image editing. Supports up to 10 reference images and 4MP output.","tagline":"Recommended default for image generation and editing.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"s/gemma-3-27b","name":"Gemma 3 27b","created_by":"google","source":"official","last_updated":"2026-06-03","family":"gemma-3","context_window":64000,"capabilities":{"streaming":true},"pricing":{"input":0.3,"output":1.2},"model_type":"chat","parameters":27,"page_url":"https://www.together.ai/models/gemma-3-27b","modalities":{"input":["text"],"output":["text"]}},{"id":"s/gemma-4-31b-it-pearl","name":"Gemma 4 31b It Pearl","created_by":"google","source":"official","family":"gemma-4","context_window":32000,"parameters":31,"page_url":"https://www.together.ai/models/gemma-4-31b-it-pearl","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.28,"output":0.86},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-17"},{"id":"s/gemma-4-31b","name":"Gemma 4 31b","created_by":"google","source":"official","family":"gemma-4","context_window":256000,"parameters":31,"page_url":"https://www.together.ai/models/gemma-4-31b","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.39,"output":0.97},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-17"},{"id":"s/glm-4-7","name":"Glm 4 7","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202000,"capabilities":{"streaming":true},"pricing":{"input":0.45,"output":2},"family":"glm-4","model_type":"chat","page_url":"https://www.together.ai/models/glm-4-7","modalities":{"input":["text"],"output":["text"]}},{"id":"s/glm-5","name":"Glm 5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202000,"capabilities":{"streaming":true},"pricing":{"input":1,"output":3.2},"family":"glm-5","model_type":"chat","page_url":"https://www.together.ai/models/glm-5","modalities":{"input":["text"],"output":["text"]}},{"id":"s/glm-51","name":"Glm 51","created_by":"zhipu","source":"official","family":"glm-51","context_window":200000,"page_url":"https://www.together.ai/models/glm-51","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.26,"output":4.4},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10"},{"id":"s/glm-52","name":"Glm 52","created_by":"zhipu","source":"official","family":"glm-52","page_url":"https://www.together.ai/models/glm-52","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.26,"output":4.4},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"s/google-veo-3-0","name":"Google Veo 3 0","created_by":"unknown","source":"official","last_updated":"2026-04-12","capabilities":{"streaming":true},"pricing":{"input":0.19,"output":1.6},"page_url":"https://www.together.ai/models/google-veo-3-0"},{"id":"s/gpt-image-1-5","name":"Gpt Image 1 5","created_by":"openai","source":"official","last_updated":"2026-06-20","family":"gpt-image-1","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0.034},"model_type":"image","page_url":"https://www.together.ai/models/gpt-image-1-5"},{"id":"s/gpt-image-2","name":"Gpt Image 2","created_by":"openai","source":"official","family":"gpt-image-2","page_url":"https://www.together.ai/models/gpt-image-2","model_type":"image","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":0.06,"output":0.053},"last_updated":"2026-06-20","description":"GPT Image 2 is our state-of-the-art image generation model for fast, high-quality image generation and editing. It supports flexible image sizes and high-fidelity image inputs. Learn more in our [image generation guide](/api/docs/guides/image-generation), or see the [pricing page](/api/docs/pricing#image-generation) and [image generation calculator](/api/docs/guides/image-generation#calculating-costs) for cost estimates.","tagline":"State-of-the-art image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"s/gpt-oss-120b","name":"Gpt Oss 120b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","parameters":120,"page_url":"https://www.together.ai/models/gpt-oss-120b","modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Most powerful open-weight model, fits into an H100 GPU","max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"s/ideogram-40","name":"Ideogram 40","created_by":"unknown","source":"official","context_window":32000,"page_url":"https://www.together.ai/models/ideogram-40","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.06},"last_updated":"2026-06-10"},{"id":"s/kimi-k2-0905","name":"Kimi K2 0905","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":1,"output":3},"family":"kimi-k2","model_type":"chat","page_url":"https://www.together.ai/models/kimi-k2-0905","modalities":{"input":["text"],"output":["text"]}},{"id":"s/kimi-k2-5","name":"Kimi K2 5","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"family":"kimi-k2","model_type":"chat","page_url":"https://www.together.ai/models/kimi-k2-5","modalities":{"input":["text"],"output":["text"]}},{"id":"s/kimi-k26","name":"Kimi K26","created_by":"moonshot","source":"official","family":"kimi-k26","context_window":256000,"page_url":"https://www.together.ai/models/kimi-k26","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.2,"output":4.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25"},{"id":"s/kimi-k27-code","name":"Kimi K27 Code","created_by":"moonshot","source":"official","family":"kimi-k27","context_window":256000,"page_url":"https://www.together.ai/models/kimi-k27-code","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.19,"output":4},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"s/lfm2-24b-a2b","name":"Lfm2 24b A2b","created_by":"liquid","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.12},"parameters":24,"active_parameters":2,"page_url":"https://www.together.ai/models/lfm2-24b-a2b","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"s/llama-4-maverick","name":"Llama 4 Maverick","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","context_window":1048000,"capabilities":{"streaming":true},"pricing":{"input":0.27,"output":0.85},"model_type":"chat","page_url":"https://www.together.ai/models/llama-4-maverick","modalities":{"input":["text"],"output":["text"]}},{"id":"s/minimax-m2-5","name":"Minimax M2 5","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":192000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":1.2},"family":"minimax","model_type":"chat","page_url":"https://www.together.ai/models/minimax-m2-5","modalities":{"input":["text"],"output":["text"]}},{"id":"s/minimax-m2-7","name":"Minimax M2 7","created_by":"minimax","source":"official","family":"minimax","page_url":"https://www.together.ai/models/minimax-m2-7","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.06,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-12"},{"id":"s/minimax-m3","name":"Minimax M3","created_by":"minimax","source":"official","family":"minimax","page_url":"https://www.together.ai/models/minimax-m3","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.06,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","context_window":1000000,"license":"proprietary","open_weight":false},{"id":"s/minimax-speech-2-6-turbo","name":"Minimax Speech 2 6 Turbo","created_by":"minimax","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"family":"minimax","model_type":"chat","page_url":"https://www.together.ai/models/minimax-speech-2-6-turbo","modalities":{"input":["text"],"output":["text"]}},{"id":"s/minimax-speech-28","name":"Minimax Speech 28","created_by":"minimax","source":"official","family":"minimax","page_url":"https://www.together.ai/models/minimax-speech-28","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-23"},{"id":"s/ministral-3-8b-instruct-2512","name":"Ministral 3 8b Instruct 2512","created_by":"mistral","source":"official","last_updated":"2026-04-10","family":"ministral","context_window":256000,"alias":"s/ministral-3-8b-instruct","capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"model_type":"chat","parameters":8,"page_url":"https://www.together.ai/models/ministral-3-8b-instruct-2512","modalities":{"input":["text"],"output":["text"]}},{"id":"s/ministral-3-8b-instruct","name":"Ministral 3 8b Instruct 2512","created_by":"mistral","source":"official","last_updated":"2026-04-10","family":"ministral","context_window":256000,"capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"snapshots":["s/ministral-3-8b-instruct-2512"],"model_type":"chat","parameters":8,"page_url":"https://www.together.ai/models/ministral-3-8b-instruct-2512","modalities":{"input":["text"],"output":["text"]}},{"id":"s/nano-banana-pro","name":"Nano Banana Pro","created_by":"unknown","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.134},"page_url":"https://www.together.ai/models/nano-banana-pro"},{"id":"s/nvidia-nemotron-3-nano-omni","name":"Nvidia Nemotron 3 Nano Omni","created_by":"nvidia","source":"official","context_window":256000,"page_url":"https://www.together.ai/models/nvidia-nemotron-3-nano-omni","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":3.75},"last_updated":"2026-06-18"},{"id":"s/nvidia-nemotron-3-super","name":"Nvidia Nemotron 3 Super","created_by":"nvidia","source":"official","last_updated":"2026-06-03","capabilities":{"streaming":true},"pricing":{"input":0.3,"output":1.2},"page_url":"https://www.together.ai/models/nvidia-nemotron-3-super"},{"id":"s/nvidia-nemotron-3-ultra","name":"Nvidia Nemotron 3 Ultra","created_by":"nvidia","source":"official","page_url":"https://www.together.ai/models/nvidia-nemotron-3-ultra","capabilities":{"streaming":true},"pricing":{"input":0.2,"output":3.6},"last_updated":"2026-06-10"},{"id":"s/nvidia-nemotron-35-asr","name":"Nvidia Nemotron 35 Asr","created_by":"nvidia","source":"official","page_url":"https://www.together.ai/models/nvidia-nemotron-35-asr","capabilities":{"streaming":true},"pricing":{"input":0.0015,"output":0.0045},"last_updated":"2026-06-10"},{"id":"s/qwen3-235b-a22b-instruct-2507-fp8","name":"Qwen3 235b A22b Instruct 2507 Fp8","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":235,"active_parameters":22,"page_url":"https://www.together.ai/models/qwen3-235b-a22b-instruct-2507-fp8","modalities":{"input":["text"],"output":["text"]}},{"id":"s/qwen3-5-397b-a17b","name":"Qwen3 5 397b A17b","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","context_window":256000,"capabilities":{"streaming":true},"pricing":{"input":0.35,"output":3.6},"model_type":"chat","parameters":397,"active_parameters":17,"page_url":"https://www.together.ai/models/qwen3-5-397b-a17b","modalities":{"input":["text"],"output":["text"]}},{"id":"s/qwen3-5-9b","name":"Qwen3 5 9b","created_by":"qwen","source":"official","last_updated":"2026-06-02","family":"qwen3","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":0.17,"output":0.25},"model_type":"chat","parameters":9,"page_url":"https://www.together.ai/models/qwen3-5-9b","modalities":{"input":["text"],"output":["text"]}},{"id":"s/qwen3-coder-next","name":"Qwen3 Coder Next","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262000,"capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":1.2},"model_type":"code","page_url":"https://www.together.ai/models/qwen3-coder-next","modalities":{"input":["text"],"output":["text"]},"description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","max_output_tokens":65536,"license":"apache-2.0","open_weight":true},{"id":"s/qwen36-plus","name":"Qwen36 Plus","created_by":"qwen","source":"official","family":"qwen36","page_url":"https://www.together.ai/models/qwen36-plus","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.5,"output":3},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"s/qwen37-max","name":"Qwen37 Max","created_by":"qwen","source":"official","family":"qwen37","page_url":"https://www.together.ai/models/qwen37-max","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.13,"output":3.75},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-29"},{"id":"s/qwen37-plus","name":"Qwen37 Plus","created_by":"qwen","source":"official","family":"qwen37","page_url":"https://www.together.ai/models/qwen37-plus","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.32,"output":1.28},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"s/rime-arcana-v3","name":"Rime Arcana V3","created_by":"unknown","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"page_url":"https://www.together.ai/models/rime-arcana-v3"},{"id":"s/rime-mist-v3-omni","name":"Rime Mist V3 Omni","created_by":"unknown","source":"official","page_url":"https://www.together.ai/models/rime-mist-v3-omni","capabilities":{"streaming":true},"last_updated":"2026-05-17"},{"id":"s/seedream-50-lite","name":"Seedream 50 Lite","created_by":"unknown","source":"official","page_url":"https://www.together.ai/models/seedream-50-lite","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.04},"last_updated":"2026-04-12","context_window":32000},{"id":"s/sora-2-pro","name":"Sora 2 Pro","created_by":"unknown","source":"official","last_updated":"2026-04-04","capabilities":{"streaming":true},"pricing":{"input":0.8,"output":3},"model_type":"video","family":"sora-2","page_url":"https://www.together.ai/models/sora-2-pro"},{"id":"s/wan-2-6-image","name":"Wan 2 6 Image","created_by":"unknown","source":"official","last_updated":"2026-06-20","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0.03},"family":"wan","model_type":"image","page_url":"https://www.together.ai/models/wan-2-6-image"},{"id":"s/wan-27","name":"Wan 27","created_by":"unknown","source":"official","family":"wan","page_url":"https://www.together.ai/models/wan-27","model_type":"video","capabilities":{"streaming":true},"pricing":{"input":3,"output":0.8},"last_updated":"2026-04-04"}]};
+export const provider: ProviderWithModels = {"id":"together","name":"Together AI","region":"US","headquarters":"San Francisco, CA","founded":2022,"url":"https://www.together.ai","api_url":"https://api.together.xyz/v1","docs_url":"https://docs.together.ai","pricing_url":"https://www.together.ai/pricing","description":"Open-source AI inference and fine-tuning platform with 100+ models.","type":"cloud","playground_url":"https://api.together.ai/playground","sdk":{"python":"together","javascript":"together-ai"},"openai_compatible":true,"free_tier":true,"github_url":"https://github.com/togethercomputer","models_url":"https://docs.together.ai/docs/serverless-models","twitter_url":"https://x.com/togethercompute","blog_url":"https://www.together.ai/blog","terms_url":"https://www.together.ai/terms-of-service","support_url":"https://www.together.ai/support","icon":"","models":[{"id":"s/cartesia-sonic-3","name":"Cartesia Sonic 3","created_by":"unknown","source":"official","page_url":"https://www.together.ai/models/cartesia-sonic-3","capabilities":{"streaming":true},"last_updated":"2026-03-27"},{"id":"s/deepseek-v3-1","name":"Deepseek V3 1","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.6,"output":1.7},"page_url":"https://www.together.ai/models/deepseek-v3-1","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"s/deepseek-v3-2-exp","name":"Deepseek V3 2 Exp","created_by":"deepseek","source":"official","last_updated":"2026-04-10","family":"deepseek","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"page_url":"https://www.together.ai/models/deepseek-v3-2-exp","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"s/deepseek-v4-flash","name":"Deepseek V4 Flash","created_by":"deepseek","source":"official","family":"deepseek","page_url":"https://www.together.ai/models/deepseek-v4-flash","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.3,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-03","description":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","context_window":1000000,"max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"s/deepseek-v4-pro","name":"Deepseek V4 Pro","created_by":"deepseek","source":"official","family":"deepseek","context_window":512000,"page_url":"https://www.together.ai/models/deepseek-v4-pro","model_type":"chat","capabilities":{"streaming":true,"json_mode":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.2,"output":3.48},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"s/flux-2-pro","name":"Flux 2 Pro","created_by":"black-forest-labs","source":"official","last_updated":"2026-04-12","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.134,"output":0.03},"model_type":"image","page_url":"https://www.together.ai/models/flux-2-pro","description":"Production-grade default model for FLUX.2 text-to-image generation and image editing. Supports up to 10 reference images and 4MP output.","tagline":"Recommended default for image generation and editing.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"s/gemma-3-27b","name":"Gemma 3 27b","created_by":"google","source":"official","last_updated":"2026-06-03","family":"gemma-3","context_window":64000,"capabilities":{"streaming":true},"pricing":{"input":0.3,"output":1.2},"model_type":"chat","parameters":27,"page_url":"https://www.together.ai/models/gemma-3-27b","modalities":{"input":["text"],"output":["text"]}},{"id":"s/gemma-4-31b-it-pearl","name":"Gemma 4 31b It Pearl","created_by":"google","source":"official","family":"gemma-4","context_window":32000,"parameters":31,"page_url":"https://www.together.ai/models/gemma-4-31b-it-pearl","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.28,"output":0.86},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-17"},{"id":"s/gemma-4-31b","name":"Gemma 4 31b","created_by":"google","source":"official","family":"gemma-4","context_window":256000,"parameters":31,"page_url":"https://www.together.ai/models/gemma-4-31b","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.39,"output":0.97},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-17"},{"id":"s/glm-4-7","name":"Glm 4 7","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202000,"capabilities":{"streaming":true},"pricing":{"input":0.45,"output":2},"family":"glm-4","model_type":"chat","page_url":"https://www.together.ai/models/glm-4-7","modalities":{"input":["text"],"output":["text"]}},{"id":"s/glm-5","name":"Glm 5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202000,"capabilities":{"streaming":true},"pricing":{"input":1,"output":3.2},"family":"glm-5","model_type":"chat","page_url":"https://www.together.ai/models/glm-5","modalities":{"input":["text"],"output":["text"]}},{"id":"s/glm-51","name":"Glm 51","created_by":"zhipu","source":"official","family":"glm-51","context_window":200000,"page_url":"https://www.together.ai/models/glm-51","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.26,"output":4.4},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10"},{"id":"s/glm-52","name":"Glm 52","created_by":"zhipu","source":"official","family":"glm-52","page_url":"https://www.together.ai/models/glm-52","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.26,"output":4.4},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29","context_window":256000},{"id":"s/google-veo-3-0","name":"Google Veo 3 0","created_by":"unknown","source":"official","last_updated":"2026-04-12","capabilities":{"streaming":true},"pricing":{"input":0.19,"output":1.6},"page_url":"https://www.together.ai/models/google-veo-3-0"},{"id":"s/gpt-image-1-5","name":"Gpt Image 1 5","created_by":"openai","source":"official","last_updated":"2026-06-29","family":"gpt-image-1","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.053,"output":0.034},"model_type":"image","page_url":"https://www.together.ai/models/gpt-image-1-5"},{"id":"s/gpt-image-2","name":"Gpt Image 2","created_by":"openai","source":"official","family":"gpt-image-2","page_url":"https://www.together.ai/models/gpt-image-2","model_type":"image","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":0.06,"output":0.053},"last_updated":"2026-06-20","description":"GPT Image 2 is our state-of-the-art image generation model for fast, high-quality image generation and editing. It supports flexible image sizes and high-fidelity image inputs. Learn more in our [image generation guide](/api/docs/guides/image-generation), or see the [pricing page](/api/docs/pricing#image-generation) and [image generation calculator](/api/docs/guides/image-generation#calculating-costs) for cost estimates.","tagline":"State-of-the-art image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"s/gpt-oss-120b","name":"Gpt Oss 120b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","parameters":120,"page_url":"https://www.together.ai/models/gpt-oss-120b","modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Most powerful open-weight model, fits into an H100 GPU","max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"s/ideogram-40","name":"Ideogram 40","created_by":"unknown","source":"official","context_window":32000,"page_url":"https://www.together.ai/models/ideogram-40","capabilities":{"streaming":true},"pricing":{"input":0.053,"output":0.06},"last_updated":"2026-06-29"},{"id":"s/kimi-k2-0905","name":"Kimi K2 0905","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":1,"output":3},"family":"kimi-k2","model_type":"chat","page_url":"https://www.together.ai/models/kimi-k2-0905","modalities":{"input":["text"],"output":["text"]}},{"id":"s/kimi-k2-5","name":"Kimi K2 5","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"family":"kimi-k2","model_type":"chat","page_url":"https://www.together.ai/models/kimi-k2-5","modalities":{"input":["text"],"output":["text"]}},{"id":"s/kimi-k26","name":"Kimi K26","created_by":"moonshot","source":"official","family":"kimi-k26","context_window":256000,"page_url":"https://www.together.ai/models/kimi-k26","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.2,"output":4.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-25"},{"id":"s/kimi-k27-code","name":"Kimi K27 Code","created_by":"moonshot","source":"official","family":"kimi-k27","context_window":256000,"page_url":"https://www.together.ai/models/kimi-k27-code","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.19,"output":4},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"s/lfm2-24b-a2b","name":"Lfm2 24b A2b","created_by":"liquid","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.12},"parameters":24,"active_parameters":2,"page_url":"https://www.together.ai/models/lfm2-24b-a2b","model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"s/llama-4-maverick","name":"Llama 4 Maverick","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","context_window":1048000,"capabilities":{"streaming":true},"pricing":{"input":0.27,"output":0.85},"model_type":"chat","page_url":"https://www.together.ai/models/llama-4-maverick","modalities":{"input":["text"],"output":["text"]}},{"id":"s/minimax-m2-5","name":"Minimax M2 5","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":192000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":1.2},"family":"minimax","model_type":"chat","page_url":"https://www.together.ai/models/minimax-m2-5","modalities":{"input":["text"],"output":["text"]}},{"id":"s/minimax-m2-7","name":"Minimax M2 7","created_by":"minimax","source":"official","family":"minimax","page_url":"https://www.together.ai/models/minimax-m2-7","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.06,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-12"},{"id":"s/minimax-m3","name":"Minimax M3","created_by":"minimax","source":"official","family":"minimax","page_url":"https://www.together.ai/models/minimax-m3","model_type":"chat","capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.06,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","context_window":1000000,"license":"proprietary","open_weight":false},{"id":"s/minimax-speech-2-6-turbo","name":"Minimax Speech 2 6 Turbo","created_by":"minimax","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"family":"minimax","model_type":"chat","page_url":"https://www.together.ai/models/minimax-speech-2-6-turbo","modalities":{"input":["text"],"output":["text"]}},{"id":"s/minimax-speech-28","name":"Minimax Speech 28","created_by":"minimax","source":"official","family":"minimax","page_url":"https://www.together.ai/models/minimax-speech-28","model_type":"chat","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-23"},{"id":"s/ministral-3-8b-instruct-2512","name":"Ministral 3 8b Instruct 2512","created_by":"mistral","source":"official","last_updated":"2026-04-10","family":"ministral","context_window":256000,"alias":"s/ministral-3-8b-instruct","capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"model_type":"chat","parameters":8,"page_url":"https://www.together.ai/models/ministral-3-8b-instruct-2512","modalities":{"input":["text"],"output":["text"]}},{"id":"s/ministral-3-8b-instruct","name":"Ministral 3 8b Instruct 2512","created_by":"mistral","source":"official","last_updated":"2026-04-10","family":"ministral","context_window":256000,"capabilities":{"streaming":true},"pricing":{"input":0.5,"output":2.8},"snapshots":["s/ministral-3-8b-instruct-2512"],"model_type":"chat","parameters":8,"page_url":"https://www.together.ai/models/ministral-3-8b-instruct-2512","modalities":{"input":["text"],"output":["text"]}},{"id":"s/nano-banana-pro","name":"Nano Banana Pro","created_by":"unknown","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.134},"page_url":"https://www.together.ai/models/nano-banana-pro"},{"id":"s/nvidia-nemotron-3-nano-omni","name":"Nvidia Nemotron 3 Nano Omni","created_by":"nvidia","source":"official","context_window":256000,"page_url":"https://www.together.ai/models/nvidia-nemotron-3-nano-omni","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":3.75},"last_updated":"2026-06-18"},{"id":"s/nvidia-nemotron-3-super","name":"Nvidia Nemotron 3 Super","created_by":"nvidia","source":"official","last_updated":"2026-06-03","capabilities":{"streaming":true},"pricing":{"input":0.3,"output":1.2},"page_url":"https://www.together.ai/models/nvidia-nemotron-3-super"},{"id":"s/nvidia-nemotron-3-ultra","name":"Nvidia Nemotron 3 Ultra","created_by":"nvidia","source":"official","page_url":"https://www.together.ai/models/nvidia-nemotron-3-ultra","capabilities":{"streaming":true},"pricing":{"input":0.2,"output":3.6},"last_updated":"2026-06-10"},{"id":"s/nvidia-nemotron-35-asr","name":"Nvidia Nemotron 35 Asr","created_by":"nvidia","source":"official","page_url":"https://www.together.ai/models/nvidia-nemotron-35-asr","capabilities":{"streaming":true},"pricing":{"input":0.0015,"output":0.0045},"last_updated":"2026-06-10"},{"id":"s/qwen3-235b-a22b-instruct-2507-fp8","name":"Qwen3 235b A22b Instruct 2507 Fp8","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":235,"active_parameters":22,"page_url":"https://www.together.ai/models/qwen3-235b-a22b-instruct-2507-fp8","modalities":{"input":["text"],"output":["text"]}},{"id":"s/qwen3-5-397b-a17b","name":"Qwen3 5 397b A17b","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen3","context_window":256000,"capabilities":{"streaming":true},"pricing":{"input":0.35,"output":3.6},"model_type":"chat","parameters":397,"active_parameters":17,"page_url":"https://www.together.ai/models/qwen3-5-397b-a17b","modalities":{"input":["text"],"output":["text"]}},{"id":"s/qwen3-5-9b","name":"Qwen3 5 9b","created_by":"qwen","source":"official","last_updated":"2026-06-02","family":"qwen3","context_window":262000,"capabilities":{"streaming":true},"pricing":{"input":0.17,"output":0.25},"model_type":"chat","parameters":9,"page_url":"https://www.together.ai/models/qwen3-5-9b","modalities":{"input":["text"],"output":["text"]}},{"id":"s/qwen3-coder-next","name":"Qwen3 Coder Next","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262000,"capabilities":{"streaming":true,"fine_tuning":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":1.2},"model_type":"code","page_url":"https://www.together.ai/models/qwen3-coder-next","modalities":{"input":["text"],"output":["text"]},"description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","max_output_tokens":65536,"license":"apache-2.0","open_weight":true},{"id":"s/qwen36-plus","name":"Qwen36 Plus","created_by":"qwen","source":"official","family":"qwen36","page_url":"https://www.together.ai/models/qwen36-plus","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.5,"output":3},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"s/qwen37-max","name":"Qwen37 Max","created_by":"qwen","source":"official","family":"qwen37","page_url":"https://www.together.ai/models/qwen37-max","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.13,"output":3.75},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-29"},{"id":"s/qwen37-plus","name":"Qwen37 Plus","created_by":"qwen","source":"official","family":"qwen37","page_url":"https://www.together.ai/models/qwen37-plus","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0.32,"output":1.28},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-18"},{"id":"s/rime-arcana-v3","name":"Rime Arcana V3","created_by":"unknown","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"page_url":"https://www.together.ai/models/rime-arcana-v3"},{"id":"s/rime-mist-v3-omni","name":"Rime Mist V3 Omni","created_by":"unknown","source":"official","page_url":"https://www.together.ai/models/rime-mist-v3-omni","capabilities":{"streaming":true},"last_updated":"2026-05-17"},{"id":"s/seedream-50-lite","name":"Seedream 50 Lite","created_by":"unknown","source":"official","page_url":"https://www.together.ai/models/seedream-50-lite","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0.04},"last_updated":"2026-04-12","context_window":32000},{"id":"s/sora-2-pro","name":"Sora 2 Pro","created_by":"unknown","source":"official","last_updated":"2026-04-04","capabilities":{"streaming":true},"pricing":{"input":0.8,"output":3},"model_type":"video","family":"sora-2","page_url":"https://www.together.ai/models/sora-2-pro"},{"id":"s/wan-2-6-image","name":"Wan 2 6 Image","created_by":"unknown","source":"official","last_updated":"2026-06-29","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.053,"output":0.03},"family":"wan","model_type":"image","page_url":"https://www.together.ai/models/wan-2-6-image"},{"id":"s/wan-27","name":"Wan 27","created_by":"unknown","source":"official","family":"wan","page_url":"https://www.together.ai/models/wan-27","model_type":"video","capabilities":{"streaming":true},"pricing":{"input":3,"output":0.8},"last_updated":"2026-04-04"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/vercel.ts b/packages/npm/src/providers/vercel.ts
index f0f2e5531..b83d47151 100644
--- a/packages/npm/src/providers/vercel.ts
+++ b/packages/npm/src/providers/vercel.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"vercel","name":"Vercel AI Gateway","region":"US","headquarters":"San Francisco, CA","founded":2015,"url":"https://vercel.com/ai-gateway","api_url":"https://gateway.ai.vercel.com/v1","docs_url":"https://vercel.com/docs/ai-gateway","pricing_url":"https://vercel.com/ai-gateway/models","description":"Vercel AI Gateway provides unified access to 200+ models through the AI SDK.","type":"aggregator","sdk":{"javascript":"ai"},"openai_compatible":true,"free_tier":false,"github_url":"https://github.com/vercel/ai","models_url":"https://sdk.vercel.ai/docs/foundations/providers-and-models","twitter_url":"https://x.com/vercel","blog_url":"https://vercel.com/blog","terms_url":"https://vercel.com/legal/terms","support_url":"https://vercel.com/help","icon":"","models":[{"id":"alibaba/qwen-3-14b","name":"qwen-3-14b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen","context_window":41000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.12,"output":0.24},"model_type":"chat","parameters":14,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen-3-235b","name":"qwen-3-235b","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.09,"output":0.1},"model_type":"chat","parameters":235,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen-3-30b","name":"qwen-3-30b","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen","context_window":41000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.12,"output":0.5},"model_type":"chat","parameters":30,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen-3-32b","name":"qwen-3-32b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","parameters":32,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen-3.6-max-preview","name":"qwen-3.6-max-preview","created_by":"qwen","source":"official","family":"qwen","context_window":240000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.3,"output":7.8},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-21","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks.","max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-235b-a22b-thinking","name":"qwen3-235b-a22b-thinking","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.23,"output":2.3},"model_type":"chat","parameters":235,"active_parameters":22,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-coder-30b-a3b","name":"qwen3-coder-30b-a3b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","parameters":30,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-coder-next","name":"qwen3-coder-next","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":1.2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","max_output_tokens":65536,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-coder-plus","name":"qwen3-coder-plus","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1,"output":5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen code models: The latest Qwen3-Coder-Plus series are code generation models built on Qwen3, featuring robust Coding Agent capabilities. They excel at tool calling and environment interaction, enabling autonomous programming with exceptional coding skills while retaining general-purpose functionality. [Usage](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/efm/model_experience_center/text?currentTab=textChat&modelId=qwen3-coder-plus) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Coder is a code generation model.","max_output_tokens":65536,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-coder","name":"qwen3-coder","created_by":"qwen","source":"official","last_updated":"2026-05-07","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.22,"output":1.8},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-embedding-0.6b","name":"qwen3-embedding-0.6b","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","context_window":32800,"model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.01,"output":0},"parameters":0.6},{"id":"alibaba/qwen3-embedding-4b","name":"qwen3-embedding-4b","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","context_window":32800,"model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"parameters":4},{"id":"alibaba/qwen3-embedding-8b","name":"qwen3-embedding-8b","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","context_window":32800,"model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.05,"output":0},"parameters":8},{"id":"alibaba/qwen3-max-preview","name":"qwen3-max-preview","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-max-thinking","name":"qwen3-max-thinking","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-max","name":"qwen3-max","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","max_output_tokens":32768,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-next-80b-a3b-instruct","name":"qwen3-next-80b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-05-07","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.15,"output":1.2},"model_type":"chat","parameters":80,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-next-80b-a3b-thinking","name":"qwen3-next-80b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-05-07","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.15,"output":1.2},"model_type":"chat","parameters":80,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-vl-235b-a22b-instruct","name":"qwen3-vl-235b-a22b-instruct","created_by":"qwen","source":"official","family":"qwen3","context_window":262100,"model_type":"chat","parameters":235,"active_parameters":22,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.88},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-vl-instruct","name":"qwen3-vl-instruct","created_by":"qwen","source":"official","last_updated":"2026-05-01","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.2,"output":0.88},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-vl-thinking","name":"qwen3-vl-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-07","family":"qwen3","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.4,"output":3.95},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3.5-flash","name":"qwen3.5-flash","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The fastest and most cost-effective model in the Qwen series, ideal for simple tasks.","max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.5-plus","name":"qwen3.5-plus","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.4,"output":2.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","max_output_tokens":64000,"license":"proprietary","open_weight":false},{"id":"alibaba/qwen3.6-27b","name":"qwen3.6-27b","created_by":"qwen","source":"official","family":"qwen3.6","context_window":256000,"model_type":"chat","parameters":27,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":3.6},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.6-plus","name":"qwen3.6-plus","created_by":"qwen","source":"official","family":"qwen3.6","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks.","max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.7-max","name":"qwen3.7-max","created_by":"qwen","source":"official","family":"qwen3.7","context_window":991000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":1.25,"output":3.75},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.7-plus","name":"qwen3.7-plus","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.32,"output":1.28},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18","license":"apache-2.0","open_weight":true},{"id":"alibaba/wan-v2.5-t2v-preview","name":"wan-v2.5-t2v-preview","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-i2v-flash","name":"wan-v2.6-i2v-flash","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-i2v","name":"wan-v2.6-i2v","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-r2v-flash","name":"wan-v2.6-r2v-flash","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-r2v","name":"wan-v2.6-r2v","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-t2v","name":"wan-v2.6-t2v","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"amazon/nova-2-lite","name":"nova-2-lite","created_by":"amazon","source":"official","last_updated":"2026-03-23","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"reasoning":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"amazon/nova-lite","name":"nova-lite","created_by":"amazon","source":"official","last_updated":"2026-03-23","context_window":300000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0.24},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"amazon/nova-micro","name":"nova-micro","created_by":"amazon","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.04,"output":0.14},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"amazon/nova-pro","name":"nova-pro","created_by":"amazon","source":"official","last_updated":"2026-03-23","context_window":300000,"capabilities":{"streaming":true},"pricing":{"input":0.8,"output":3.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"amazon/titan-embed-text-v2","name":"titan-embed-text-v2","created_by":"amazon","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0}},{"id":"anthropic/claude-3-haiku","name":"claude-3-haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true},"pricing":{"input":0.25,"output":1.25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"proprietary","open_weight":false},{"id":"anthropic/claude-3-opus","name":"claude-3-opus","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true},"pricing":{"input":15,"output":75},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"license":"proprietary","open_weight":false},{"id":"anthropic/claude-3.5-haiku","name":"claude-3.5-haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.8,"output":4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-3.5-sonnet-20240620","name":"claude-3.5-sonnet-20240620","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":200000,"alias":"anthropic/claude-3.5-sonnet","capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-3.5-sonnet","name":"claude-3.5-sonnet-20240620","created_by":"anthropic","source":"official","last_updated":"2026-03-26","family":"claude-sonnet","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":3,"output":15},"snapshots":["anthropic/claude-3.5-sonnet-20240620"],"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-3.7-sonnet","name":"claude-3.7-sonnet","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-fable-5","name":"claude-fable-5","created_by":"anthropic","source":"official","family":"claude","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":50},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10","description":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","tagline":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","max_output_tokens":128000,"license":"proprietary","open_weight":false},{"id":"anthropic/claude-haiku-4.5","name":"claude-haiku-4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1,"output":5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.1","name":"claude-opus-4.1","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":15,"output":75},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.5","name":"claude-opus-4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.6","name":"claude-opus-4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.7","name":"claude-opus-4.7","created_by":"anthropic","source":"official","family":"claude-opus","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"anthropic/claude-opus-4.8","name":"claude-opus-4.8","created_by":"anthropic","source":"official","family":"claude-opus","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"anthropic/claude-opus-4","name":"claude-opus-4","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":15,"output":75},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-sonnet-4.5","name":"claude-sonnet-4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-sonnet-4.6","name":"claude-sonnet-4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-sonnet-4","name":"claude-sonnet-4","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"arcee-ai/trinity-large-preview","name":"trinity-large-preview","created_by":"arcee","source":"official","last_updated":"2026-03-23","context_window":131000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.25,"output":1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"arcee-ai/trinity-large-thinking","name":"trinity-large-thinking","created_by":"arcee","source":"official","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.25,"output":0.9},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"arcee-ai/trinity-mini","name":"trinity-mini","created_by":"arcee","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true},"pricing":{"input":0.04,"output":0.15},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"bfl/flux-2-flex","name":"flux-2-flex","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"FLUX.2 variant with adjustable steps and guidance, tuned for typography and editing workflows. Up to 8 references via API.","tagline":"Flexible FLUX.2 for typography and editing.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-klein-4b","name":"flux-2-klein-4b","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"fine_tuning":true},"pricing":{"input":0,"output":0},"model_type":"image","parameters":4,"description":"Lightweight 4B-parameter FLUX.2 flow model for sub-second inference; ~13 GB VRAM. Open base weights under Apache 2.0.","tagline":"Fastest, most lightweight FLUX.2.","license":"apache-2.0","open_weight":true,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-klein-9b","name":"flux-2-klein-9b","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"fine_tuning":true},"pricing":{"input":0,"output":0},"model_type":"image","parameters":9,"description":"Balanced 9B FLUX.2 flow model with 8B text embedder; sub-second inference at ~24 GB VRAM. Open base weights under FLUX non-commercial license.","tagline":"Balanced quality and speed for [klein].","license":"flux-1-dev-non-commercial","open_weight":true,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-max","name":"flux-2-max","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","context_window":67300,"capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Highest-quality FLUX.2 model with web grounding search, up to 8 references via API and 10 in playground.","tagline":"Highest quality FLUX.2.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-pro","name":"flux-2-pro","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","context_window":67300,"capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Production-grade default model for FLUX.2 text-to-image generation and image editing. Supports up to 10 reference images and 4MP output.","tagline":"Recommended default for image generation and editing.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-kontext-max","name":"flux-kontext-max","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Higher-quality variant of Kontext with stronger editing fidelity. Legacy; superseded by FLUX.2 [max] for new projects.","tagline":"Legacy FLUX.1 Kontext [max].","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-kontext-pro","name":"flux-kontext-pro","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Context-aware image editing and text-to-image. Legacy FLUX.1 line; BFL recommends FLUX.2 [pro] for new projects.","tagline":"Legacy FLUX.1 Kontext [pro].","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-pro-1.0-fill","name":"flux-pro-1.0-fill","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Text-driven inpainting and outpainting using an input image and mask.","tagline":"Inpainting and outpainting for FLUX.1.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-pro-1.1-ultra","name":"flux-pro-1.1-ultra","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"FLUX 1.1 [pro] with up to 4MP resolution and optional Raw mode for photographic authenticity. Supports image-to-image.","tagline":"FLUX 1.1 [pro] up to 4MP with raw mode.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-pro-1.1","name":"flux-pro-1.1","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Fast and reliable text-to-image baseline with strong prompt adherence. Sometimes called FLUX1.1 [pro].","tagline":"Fast and reliable FLUX 1.1 baseline.","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seed-1.6","name":"seed-1.6","created_by":"bytedance","source":"official","last_updated":"2026-03-23","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":0.25,"output":2},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Seed1.6 integrates multimodal capabilities, with \"adaptive thinking\" balancing model performance with reasoning capability.","tagline":"Seed1.6 integrates multimodal capabilities, with \"adaptive thinking\" balancing model performance with reasoning capability."},{"id":"bytedance/seed-1.8","name":"seed-1.8","created_by":"bytedance","source":"official","last_updated":"2026-03-23","context_window":256000,"capabilities":{"streaming":true,"vision":true,"reasoning":true},"pricing":{"input":0.25,"output":2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"A general-purpose agent model capable of efficiently and accurately completing tasks in complex workflows.","tagline":"A general-purpose agent model capable of efficiently and accurately completing tasks in complex workflows."},{"id":"bytedance/seedance-2.0-fast","name":"seedance-2.0-fast","created_by":"bytedance","source":"official","model_type":"video","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16"},{"id":"bytedance/seedance-2.0","name":"seedance-2.0","created_by":"bytedance","source":"official","model_type":"video","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16","description":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","tagline":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","modalities":{"input":["text","audio"],"output":["video"]}},{"id":"bytedance/seedance-v1.0-lite-i2v","name":"seedance-v1.0-lite-i2v","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.0-lite-t2v","name":"seedance-v1.0-lite-t2v","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.0-pro-fast","name":"seedance-v1.0-pro-fast","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.0-pro","name":"seedance-v1.0-pro","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.5-pro","name":"seedance-v1.5-pro","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedream-4.0","name":"seedream-4.0","created_by":"bytedance","source":"official","model_type":"image","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-12","description":"A new-generation image creation model can flexibly handle complex multimodal tasks, producing high-definition images at up to 4K resolution.","tagline":"A new-generation image creation model can flexibly handle complex multimodal tasks, producing high-definition images at up to 4K resolution.","modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seedream-4.5","name":"seedream-4.5","created_by":"bytedance","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-12","description":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","tagline":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seedream-5.0-lite","name":"seedream-5.0-lite","created_by":"bytedance","source":"official","model_type":"image","capabilities":{"streaming":true,"vision":true,"reasoning":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-12","description":"A unified multimodal image generation model endowed with deep thinking and online search capabilities.","tagline":"A unified multimodal image generation model endowed with deep thinking and online search capabilities.","modalities":{"input":["text"],"output":["image"]}},{"id":"cohere/command-a","name":"command-a","created_by":"cohere","source":"official","last_updated":"2026-03-23","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":2.5,"output":10},"family":"command-a","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"cohere/embed-v4.0","name":"embed-v4.0","created_by":"cohere","source":"official","last_updated":"2026-06-18","model_type":"embed","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.12,"output":0},"family":"embed","context_window":128000,"description":"A model that allows for text and images to be classified or turned into embeddings","tagline":"Text, Images, Mixed texts/images (i.e.","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"cohere/rerank-v3.5","name":"rerank-v3.5","created_by":"cohere","source":"official","family":"rerank","context_window":4100,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-14","description":"A model that allows for re-ranking English Language documents and semi-structured data (JSON). This model has a context length of 4096 tokens.","tagline":"Unique per deployment","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"cohere/rerank-v4-fast","name":"rerank-v4-fast","created_by":"cohere","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16"},{"id":"cohere/rerank-v4-pro","name":"rerank-v4-pro","created_by":"cohere","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16"},{"id":"deepseek/deepseek-r1","name":"deepseek-r1","created_by":"deepseek","source":"official","last_updated":"2026-04-08","family":"deepseek-r1","context_window":160000,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":1.35,"output":5.4},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.1-terminus","name":"deepseek-v3.1-terminus","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.27,"output":1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.1","name":"deepseek-v3.1","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.21,"output":0.79},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.2-thinking","name":"deepseek-v3.2-thinking","created_by":"deepseek","source":"official","last_updated":"2026-05-01","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":0.28,"output":0.42},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.2","name":"deepseek-v3.2","created_by":"deepseek","source":"official","last_updated":"2026-05-01","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"reasoning":true},"pricing":{"input":0.28,"output":0.42},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3","name":"deepseek-v3","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.27,"output":1.12},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v4-flash","name":"deepseek-v4-flash","created_by":"deepseek","source":"official","family":"deepseek","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"fine_tuning":true},"pricing":{"input":0.14,"output":0.28},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","description":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"deepseek/deepseek-v4-pro","name":"deepseek-v4-pro","created_by":"deepseek","source":"official","family":"deepseek","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"fine_tuning":true},"pricing":{"input":0.43,"output":0.87},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","description":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"google/gemini-2.0-flash-lite","name":"gemini-2.0-flash-lite","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"json_mode":true},"pricing":{"input":0.07,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":8192,"knowledge_cutoff":"2024-08","license":"proprietary","open_weight":false},{"id":"google/gemini-2.0-flash","name":"gemini-2.0-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"json_mode":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":8192,"knowledge_cutoff":"2024-08","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-image","name":"gemini-2.5-flash-image","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":32800,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency. Gemini 2.5 Flash Image, also known as Nano Banana, is best for high-volume generation, conversational image editing, and low-latency creative workflows that require native multimodal understanding.","tagline":"Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency.","max_output_tokens":32768,"knowledge_cutoff":"2025-06","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite-preview-09-2025","name":"gemini-2.5-flash-lite-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite","name":"gemini-2.5-flash-lite","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 2.5 Flash-Lite is best for high-volume classification, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-preview-09-2025","name":"gemini-2.5-flash-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash","name":"gemini-2.5-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our best model in terms of price-performance, offering well-rounded capabilities. 2.5 Flash is best for large scale processing, low-latency, high volume tasks that require thinking, and agentic use cases.","tagline":"Our best model in terms of price-performance, offering well-rounded capabilities.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-pro","name":"gemini-2.5-pro","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","tagline":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-flash","name":"gemini-3-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemini-3-pro-image","name":"gemini-3-pro-image","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","context_window":65500,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"max_output_tokens":32768,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-preview","name":"gemini-3-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image-preview","name":"gemini-3.1-flash-image-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","context_window":131100,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"max_output_tokens":32768,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image","name":"gemini-3.1-flash-image","created_by":"google","source":"official","family":"gemini-3.1","context_window":131100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10","max_output_tokens":32768,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite-preview","name":"gemini-3.1-flash-lite-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 3.1 Flash-Lite is best for high-volume agentic tasks, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite","name":"gemini-3.1-flash-lite","created_by":"google","source":"official","family":"gemini-3.1","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-10","description":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary constraints.","tagline":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary con","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-pro-preview","name":"gemini-3.1-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.5-flash","name":"gemini-3.5-flash","created_by":"google","source":"official","family":"gemini-3.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.5,"output":9},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-embedding-001","name":"gemini-embedding-001","created_by":"google","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.15,"output":0},"family":"gemini-embedding","context_window":2048,"license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemini-embedding-2","name":"gemini-embedding-2","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-embedding","model_type":"embed","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.2,"output":0},"context_window":8192,"license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-4-26b-a4b-it","name":"gemma-4-26b-a4b-it","created_by":"google","source":"official","family":"gemma-4","context_window":262100,"model_type":"chat","parameters":26,"active_parameters":4,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.13,"output":0.4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"google/gemma-4-31b-it","name":"gemma-4-31b-it","created_by":"google","source":"official","family":"gemma-4","context_window":262100,"model_type":"chat","parameters":31,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.14,"output":0.4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"google/imagen-4.0-fast-generate-001","name":"imagen-4.0-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","family":"imagen-4.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"google/imagen-4.0-generate-001","name":"imagen-4.0-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","family":"imagen-4.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"google/imagen-4.0-ultra-generate-001","name":"imagen-4.0-ultra-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","family":"imagen-4.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"google/text-embedding-005","name":"text-embedding-005","created_by":"google","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0},"family":"text-embedding"},{"id":"google/text-multilingual-embedding-002","name":"text-multilingual-embedding-002","created_by":"google","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0}},{"id":"google/veo-3.0-fast-generate-001","name":"veo-3.0-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]}},{"id":"google/veo-3.0-generate-001","name":"veo-3.0-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]}},{"id":"google/veo-3.1-fast-generate-001","name":"veo-3.1-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.1"},{"id":"google/veo-3.1-generate-001","name":"veo-3.1-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.1"},{"id":"inception/mercury-2","name":"mercury-2","created_by":"inception","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.25,"output":0.75},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Mercury is the first diffusion large language model (dLLM). Applying a breakthrough discrete diffusion approach, the model runs 5-10x faster than even speed optimized models like Claude 3.5 Haiku and GPT-4o Mini while matching their performance. Mercury's speed means that users can stay in the flow, enjoying rapid chat-based iteration and responsive completion suggestions. On Copilot Arena, Mercury ranks 1st in speed and ties for 2nd in quality.","tagline":"The fastest reasoning dLLM.","max_output_tokens":50000,"license":"proprietary","open_weight":false},{"id":"inception/mercury-coder-small","name":"mercury-coder-small","created_by":"inception","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.25,"output":1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"interfaze/interfaze-beta","name":"interfaze-beta","created_by":"interfaze","source":"official","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":1.5,"output":3.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-05-01"},{"id":"klingai/kling-v2.5-turbo-i2v","name":"kling-v2.5-turbo-i2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.5-turbo-t2v","name":"kling-v2.5-turbo-t2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.6-i2v","name":"kling-v2.6-i2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.6-motion-control","name":"kling-v2.6-motion-control","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.6-t2v","name":"kling-v2.6-t2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v3.0-i2v","name":"kling-v3.0-i2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v3.0-motion-control","name":"kling-v3.0-motion-control","created_by":"klingai","source":"official","model_type":"video","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"klingai/kling-v3.0-t2v","name":"kling-v3.0-t2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"kwaipilot/kat-coder-pro-v1","name":"kat-coder-pro-v1","created_by":"kwaipilot","source":"official","last_updated":"2026-06-20","context_window":256000,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"kwaipilot/kat-coder-pro-v2","name":"kat-coder-pro-v2","created_by":"kwaipilot","source":"official","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"meituan/longcat-flash-chat","name":"longcat-flash-chat","created_by":"meituan","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0,"output":0},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"meituan/longcat-flash-thinking-2601","name":"longcat-flash-thinking-2601","created_by":"meituan","source":"official","last_updated":"2026-03-23","context_window":32800,"alias":"meituan/longcat-flash-thinking","capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":0,"output":0},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"meituan/longcat-flash-thinking","name":"longcat-flash-thinking-2601","created_by":"meituan","source":"official","last_updated":"2026-03-26","context_window":32800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0,"output":0},"snapshots":["meituan/longcat-flash-thinking-2601"],"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.1-70b","name":"llama-3.1-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.72,"output":0.72},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.1-8b","name":"llama-3.1-8b","created_by":"meta","source":"official","last_updated":"2026-06-02","family":"llama-3.1","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.02,"output":0.05},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-11b","name":"llama-3.2-11b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.16,"output":0.16},"model_type":"chat","parameters":11,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-1b","name":"llama-3.2-1b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","parameters":1,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-3b","name":"llama-3.2-3b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","parameters":3,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-90b","name":"llama-3.2-90b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.72,"output":0.72},"model_type":"chat","parameters":90,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.3-70b","name":"llama-3.3-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.3","context_window":128000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.59,"output":0.72},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-4-maverick","name":"llama-4-maverick","created_by":"meta","source":"official","last_updated":"2026-04-07","family":"llama-4","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.24,"output":0.97},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-4-scout","name":"llama-4-scout","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.17,"output":0.66},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"minimax/minimax-m2.1-lightning","name":"minimax-m2.1-lightning","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":2.4},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"minimax/minimax-m2.1","name":"minimax-m2.1","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.3,"output":1.2},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","tagline":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5-highspeed","name":"minimax-m2.5-highspeed","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":2.4},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"M2.5 highspeed: Same performance, faster and more agile","tagline":"M2.5 highspeed: Same performance, faster and more agile","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5","name":"minimax-m2.5","created_by":"minimax","source":"official","last_updated":"2026-06-02","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.07,"output":0.57},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Peak Performance. Ultimate Value. Master the Complex","tagline":"Peak Performance.","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.7-highspeed","name":"minimax-m2.7-highspeed","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":2.4},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"M2.7 Highspeed: Same performance, faster and more agile","tagline":"M2.7 Highspeed: Same performance, faster and more agile","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2.7","name":"minimax-m2.7","created_by":"minimax","source":"official","last_updated":"2026-06-02","context_window":204800,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.15,"output":0.6},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2","name":"minimax-m2","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":205000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"model_type":"chat","family":"minimax","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Agentic capabilities, Advanced reasoning","tagline":"Agentic capabilities, Advanced reasoning","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m3","name":"minimax-m3","created_by":"minimax","source":"official","family":"minimax","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","license":"proprietary","open_weight":false},{"id":"mistral/codestral-embed","name":"codestral-embed","created_by":"mistral","source":"official","last_updated":"2026-03-21","family":"codestral","model_type":"embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0},"context_window":8000,"license":"mnpl","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]}},{"id":"mistral/codestral","name":"codestral","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"codestral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.3,"output":0.9},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our first cutting-edge language model for coding and FIM, released under the MNPL license.","tagline":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","license":"mnpl","open_weight":true},{"id":"mistral/devstral-2","name":"devstral-2","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.4,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/devstral-small-2","name":"devstral-small-2","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/devstral-small","name":"devstral-small","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","license":"apache-2.0","open_weight":true},{"id":"mistral/magistral-medium","name":"magistral-medium","created_by":"mistral","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true},"pricing":{"input":2,"output":5},"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"description":"Our first frontier-class reasoning model released June 2025.","tagline":"Our frontier-class multimodal reasoning model update of September 2025.","license":"proprietary","open_weight":false},{"id":"mistral/magistral-small","name":"magistral-small","created_by":"mistral","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"description":"Our small reasoning model released July 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","license":"apache-2.0","open_weight":true},{"id":"mistral/ministral-14b","name":"ministral-14b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.2},"model_type":"chat","parameters":14,"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"mistral/ministral-3b","name":"ministral-3b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"mistral/ministral-8b","name":"ministral-8b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"mistral/mistral-embed","name":"mistral-embed","created_by":"mistral","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0},"family":"mistral-embed","context_window":8000,"license":"apache-2.0","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]}},{"id":"mistral/mistral-large-3","name":"mistral-large-3","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-large","context_window":256000,"capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/mistral-medium-3.5","name":"mistral-medium-3.5","created_by":"mistral","source":"official","family":"mistral-medium","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.5,"output":7.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23"},{"id":"mistral/mistral-medium","name":"mistral-medium","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-medium","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.4,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our first SOTA enterprise grade model.","tagline":"Our frontier-class multimodal model released August 2025.","license":"proprietary","open_weight":false},{"id":"mistral/mistral-nemo","name":"mistral-nemo","created_by":"mistral","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","family":"mistral-nemo","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our best multilingual open source model released July 2024.","tagline":"Our best multilingual open source model released July 2024.","license":"apache-2.0","open_weight":true},{"id":"mistral/mistral-small","name":"mistral-small","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"An enterprise grade small model for efficiency.","tagline":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model.","license":"apache-2.0","open_weight":true},{"id":"mistral/mixtral-8x22b-instruct","name":"mixtral-8x22b-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","context_window":65500,"capabilities":{"streaming":true},"pricing":{"input":1.2,"output":1.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/pixtral-12b","name":"pixtral-12b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"pixtral","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"A 12B model with image understanding capabilities in addition to text.","tagline":"A 12B model with image understanding capabilities in addition to text.","license":"apache-2.0","open_weight":true},{"id":"mistral/pixtral-large","name":"pixtral-large","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"pixtral","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"mnpl","open_weight":true},{"id":"moonshotai/kimi-k2-0905","name":"kimi-k2-0905","created_by":"moonshot","source":"official","last_updated":"2026-04-16","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.6,"output":2.5},"family":"kimi-k2","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"moonshotai/kimi-k2-thinking-turbo","name":"kimi-k2-thinking-turbo","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":1.15,"output":8},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Deep reasoning with high speed output.","tagline":"Deep reasoning with high speed output.","max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2-thinking","name":"kimi-k2-thinking","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":2.5},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Long-term thinking, multi-step tool usage, complex problem solving.","tagline":"Long-term thinking, multi-step tool usage, complex problem solving.","max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2-turbo","name":"kimi-k2-turbo","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":1.15,"output":8},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2.5","name":"kimi-k2.5","created_by":"moonshot","source":"official","last_updated":"2026-06-18","context_window":262100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":3},"family":"kimi-k2.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","tagline":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.6","name":"kimi-k2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.95,"output":4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"moonshotai/kimi-k2.7-code-highspeed","name":"kimi-k2.7-code-highspeed","created_by":"moonshot","source":"official","family":"kimi-k2.7","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.9,"output":8},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"moonshotai/kimi-k2.7-code","name":"kimi-k2.7-code","created_by":"moonshot","source":"official","family":"kimi-k2.7","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.74,"output":3.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"moonshotai/kimi-k2","name":"kimi-k2","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.57,"output":2.3},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"morph/morph-v3-fast","name":"morph-v3-fast","created_by":"morph","source":"official","last_updated":"2026-03-23","context_window":81900,"capabilities":{"streaming":true},"pricing":{"input":0.8,"output":1.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"morph/morph-v3-large","name":"morph-v3-large","created_by":"morph","source":"official","last_updated":"2026-03-23","context_window":81900,"capabilities":{"streaming":true},"pricing":{"input":0.9,"output":1.9},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"nemotron-3-nano-30b-a3b","created_by":"nvidia","source":"official","last_updated":"2026-03-23","context_window":262100,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":0.05,"output":0.24},"model_type":"chat","parameters":30,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"nemotron-3-super-120b-a12b","created_by":"nvidia","source":"official","context_window":256000,"model_type":"chat","parameters":120,"active_parameters":12,"capabilities":{"streaming":true},"pricing":{"input":0.15,"output":0.65},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-04-01"},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"nemotron-3-ultra-550b-a55b","created_by":"nvidia","source":"official","context_window":1000000,"model_type":"chat","parameters":550,"active_parameters":55,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.37,"output":1.08},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-nano-12b-v2-vl","name":"nemotron-nano-12b-v2-vl","created_by":"nvidia","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-nano-9b-v2","name":"nemotron-nano-9b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.06,"output":0.23},"model_type":"chat","parameters":9,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-3.5-turbo-instruct","name":"gpt-3.5-turbo-instruct","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-3.5","context_window":8200,"capabilities":{"streaming":true,"fine_tuning":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"reasoning":false},"pricing":{"input":1.5,"output":2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Similar capabilities as GPT-3 era models. Compatible with legacy Completions endpoint and not Chat Completions.","tagline":"An older model only compatible with the legacy Completions endpoint","max_output_tokens":4096,"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-3.5-turbo","name":"gpt-3.5-turbo","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-3.5","context_window":16400,"capabilities":{"streaming":true,"fine_tuning":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"reasoning":false,"batch":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-3.5 Turbo models can understand and generate natural language or code and have been optimized for chat using the Chat Completions API but work well for non-chat tasks as well. As of July 2024, use gpt-4o-mini in place of GPT-3.5 Turbo, as it is cheaper, more capable, multimodal, and just as fast. GPT-3.5 Turbo is still available for use in the API.","tagline":"Legacy GPT model for cheaper chat and non-chat tasks","max_output_tokens":4096,"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-4-turbo","name":"gpt-4-turbo","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4-turbo","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":10,"output":30},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4 Turbo is the next generation of GPT-4, an older high-intelligence GPT model. It was designed to be a cheaper, better version of GPT-4. Today, we recommend using a newer model like GPT-4o.","tagline":"An older high-intelligence GPT model","max_output_tokens":4096,"knowledge_cutoff":"2023-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-mini","name":"gpt-4.1-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":0.4,"output":1.6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4.1 mini excels at instruction following and tool calling. It features a","tagline":"Smaller, faster version of GPT-4.1","max_output_tokens":32768,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-nano","name":"gpt-4.1-nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4.1 nano excels at instruction following and tool calling. It features a","tagline":"Fastest, most cost-efficient version of GPT-4.1","max_output_tokens":32768,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1","name":"gpt-4.1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":2,"output":8},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4.1 excels at instruction following and tool calling, with broad","tagline":"Smartest non-reasoning model","max_output_tokens":32768,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini-search-preview","name":"gpt-4o-mini-search-preview","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4o","context_window":128000,"capabilities":{"streaming":true,"vision":true,"structured_output":true,"json_mode":true,"tool_call":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-4o mini Search Preview is a specialized model trained to understand and execute [web search](/api/docs/guides/tools-web-search?api-mode=chat) queries with the Chat Completions API. In addition to token fees, web search queries have a fee per tool call. Learn more in the [pricing](/api/docs/pricing) page.","tagline":"Fast, affordable small model for web search","max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini","name":"gpt-4o-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4o","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4o mini (“o” for “omni”) is a fast, affordable small model for focused tasks.","tagline":"Fast, affordable small model for focused tasks","max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o","name":"gpt-4o","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4o","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":false,"structured_output":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4o (“o” for “omni”) is our versatile, high-intelligence flagship model.","tagline":"Fast, intelligent, flexible GPT model","max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-chat","name":"gpt-5-chat","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5-codex","name":"gpt-5-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5 optimized for agentic coding in Codex","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-mini","name":"gpt-5-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.25,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 mini is a faster, more cost-efficient version of GPT-5. It's great for","tagline":"Near-frontier intelligence for cost sensitive, low latency, high volume workloads","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-nano","name":"gpt-5-nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.05,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for","tagline":"Fastest, most cost-efficient version of GPT-5","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-pro","name":"gpt-5-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":120},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex-max","name":"gpt-5.1-codex-max","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT‑5.1-Codex-Max is purpose-built for agentic coding.","tagline":"A version of GPT-5.1-codex optimized for long running tasks.","max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex-mini","name":"gpt-5.1-codex-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":0.25,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1-Codex mini is a smaller, more cost-effective, less-capable version of GPT-5.1-Codex.","tagline":"Smaller, more cost-effective, less-capable version of GPT-5.1-Codex","max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex","name":"gpt-5.1-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5.1 optimized for agentic coding in Codex.","max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-instant","name":"gpt-5.1-instant","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.1-thinking","name":"gpt-5.1-thinking","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.2-chat","name":"gpt-5.2-chat","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.2-codex","name":"gpt-5.2-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2-Codex is an upgraded version of GPT-5.2 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.2-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.2-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"Our most intelligent coding model optimized for long-horizon, agentic coding tasks.","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2-pro","name":"gpt-5.2-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false},"pricing":{"input":21,"output":168},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2","name":"gpt-5.2","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.3-chat","name":"gpt-5.3-chat","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.3-codex","name":"gpt-5.3-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.3-Codex is optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.3-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.3-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"The most capable agentic coding model to date.","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-mini","name":"gpt-5.4-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 mini brings the strengths of GPT-5.4 to a faster, more efficient","tagline":"Our strongest mini model yet for coding, computer use, and subagents","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-nano","name":"gpt-5.4-nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 nano is designed for tasks where speed and cost matter most like","tagline":"Our cheapest GPT-5.4-class model for simple high-volume tasks","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-pro","name":"gpt-5.4-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":1100000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false},"pricing":{"input":30,"output":180},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4","name":"gpt-5.4","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":1100000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 is our frontier model for complex professional work.","tagline":"A more affordable model for coding and professional work.","max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5-pro","name":"gpt-5.5-pro","created_by":"openai","source":"official","family":"gpt-5.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":30,"output":180},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5","name":"gpt-5.5","created_by":"openai","source":"official","family":"gpt-5.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":5,"output":30},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","description":"GPT-5.5 is our newest frontier model for the most complex professional work.","tagline":"A new class of intelligence for coding and professional work.","max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5","name":"gpt-5","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-image-1-mini","name":"gpt-image-1-mini","created_by":"openai","source":"official","last_updated":"2026-03-21","family":"gpt-image-1","capabilities":{"streaming":true},"pricing":{"input":2,"output":8},"model_type":"image","description":"A cost-efficient version of GPT Image 1. It is a natively multimodal language model that accepts both text and image inputs, and produces image outputs.","tagline":"A cost-efficient version of GPT Image 1","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"openai/gpt-image-1.5","name":"gpt-image-1.5","created_by":"openai","source":"official","last_updated":"2026-03-21","family":"gpt-image-1.5","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":5,"output":32},"model_type":"image","description":"GPT Image 1.5 is our previous image generation model, with better instruction following and adherence to prompts. Learn more in our [GPT Image 1.5 usage guide](/api/docs/guides/image-generation).","tagline":"Our previous image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image","text"]}},{"id":"openai/gpt-image-1","name":"gpt-image-1","created_by":"openai","source":"official","last_updated":"2026-03-21","family":"gpt-image-1","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":5,"output":40},"model_type":"image","description":"GPT Image 1 is a natively multimodal language model that accepts both text and image inputs, and produces image outputs.","tagline":"Our previous image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"openai/gpt-image-2","name":"gpt-image-2","created_by":"openai","source":"official","family":"gpt-image-2","model_type":"image","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":5,"output":30},"last_updated":"2026-04-22","description":"GPT Image 2 is our state-of-the-art image generation model for fast, high-quality image generation and editing. It supports flexible image sizes and high-fidelity image inputs. Learn more in our [image generation guide](/api/docs/guides/image-generation), or see the [pricing page](/api/docs/pricing#image-generation) and [image generation calculator](/api/docs/guides/image-generation#calculating-costs) for cost estimates.","tagline":"State-of-the-art image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"pricing":{"input":0.35,"output":0.75},"model_type":"chat","parameters":120,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Most powerful open-weight model, fits into an H100 GPU","max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"pricing":{"input":0.07,"output":0.3},"model_type":"chat","parameters":20,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Medium-sized open-weight model for low latency","max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-safeguard-20b","name":"gpt-oss-safeguard-20b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.07,"output":0.3},"model_type":"chat","parameters":20,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/o1","name":"o1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o1","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":60},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The o1 series of models are trained with reinforcement learning to perform complex reasoning. o1 models think before they answer, producing a long internal chain of thought before responding to the user.","tagline":"Previous full o-series reasoning model","max_output_tokens":100000,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-deep-research","name":"o3-deep-research","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false,"batch":true},"pricing":{"input":10,"output":40},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o3-deep-research is our most advanced model for deep research,","tagline":"Our most powerful deep research model","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3-mini","name":"o3-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.1,"output":4.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o3-mini is our newest small reasoning model, providing high intelligence at the same cost and latency targets of o1-mini. o3-mini supports key developer features, like Structured Outputs, function calling, and Batch API.","tagline":"A small model alternative to o3","max_output_tokens":100000,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-pro","name":"o3-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":20,"output":80},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The o-series of models are trained with reinforcement learning to think","tagline":"Version of o3 with more compute for better responses","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3","name":"o3","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2,"output":8},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o3 is a well-rounded and powerful model across domains. It sets a new","tagline":"Reasoning model for complex tasks, succeeded by GPT-5","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o4-mini","name":"o4-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o4","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"batch":true},"pricing":{"input":1.1,"output":4.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o4-mini is our latest small o-series model. It's optimized for fast,","tagline":"Fast, cost-efficient reasoning model, succeeded by GPT-5 mini","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/text-embedding-3-large","name":"text-embedding-3-large","created_by":"openai","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.13,"output":0},"family":"text-embedding","description":"text-embedding-3-large is our most capable embedding model for both english and non-english tasks.","tagline":"Most capable embedding model","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/text-embedding-3-small","name":"text-embedding-3-small","created_by":"openai","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"family":"text-embedding","description":"text-embedding-3-small is our improved, more performant version of our ada embedding model.","tagline":"Small embedding model","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/text-embedding-ada-002","name":"text-embedding-ada-002","created_by":"openai","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.1,"output":0},"family":"text-embedding","description":"text-embedding-ada-002 is our improved, more performant version of our ada embedding model.","tagline":"Older embedding model","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"perplexity/sonar-pro","name":"sonar-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0,"output":0},"family":"sonar","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"perplexity/sonar-reasoning-pro","name":"sonar-reasoning-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","context_window":127000,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":0,"output":0},"model_type":"chat","family":"sonar","modalities":{"input":["text"],"output":["text"]}},{"id":"perplexity/sonar-reasoning","name":"sonar-reasoning","created_by":"perplexity","source":"official","last_updated":"2026-03-21","context_window":127000,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":1,"output":5},"model_type":"reasoning","family":"sonar"},{"id":"perplexity/sonar","name":"sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","context_window":127000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0,"output":0},"family":"sonar","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"prime-intellect/intellect-3","name":"intellect-3","created_by":"prime-intellect","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":1.1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"prodia/flux-fast-schnell","name":"flux-fast-schnell","created_by":"prodia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image"},{"id":"recraft/recraft-v2","name":"recraft-v2","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"First-generation Recraft raster model.","tagline":"Legacy Recraft V2 raster.","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"recraft/recraft-v3","name":"recraft-v3","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Previously codenamed Red Panda; raster model with strong text rendering and brand-style preservation. Used for image-to-image, inpainting, and background editing endpoints.","tagline":"Recraft V3 raster (formerly Red Panda).","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"recraft/recraft-v4-pro","name":"recraft-v4-pro","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"High-fidelity 2048x2048 raster tier of Recraft V4 with improved anatomy and realism for print-grade output.","tagline":"Print-grade raster Recraft V4.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"recraft/recraft-v4.1-pro","name":"recraft-v4.1-pro","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4.1-utility-pro","name":"recraft-v4.1-utility-pro","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4.1-utility","name":"recraft-v4.1-utility","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4.1","name":"recraft-v4.1","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4","name":"recraft-v4","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Latest Recraft raster model with strong design sensibility, readable text, and consistent composition. Default model in the API.","tagline":"Design-forward raster image model.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"stepfun/step-3.5-flash","name":"step-3.5-flash","created_by":"stepfun","source":"official","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":true},"pricing":{"input":0.09,"output":0.3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02","description":"Flagship reasoning model with 256k context, MoE architecture (199B total, ~11B active). Optimized for agentic tasks, tool calling, planning, math, and coding.","tagline":"Frontier reasoning and agentic capabilities at low cost.","license":"apache-2.0","open_weight":true},{"id":"stepfun/step-3.7-flash","name":"step-3.7-flash","created_by":"stepfun","source":"official","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":1.15},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"voyage/rerank-2.5-lite","name":"rerank-2.5-lite","created_by":"voyage","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/rerank-2.5","name":"rerank-2.5","created_by":"voyage","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0.05,"output":0},"last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-3-large","name":"voyage-3-large","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.18,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-3.5-lite","name":"voyage-3.5-lite","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-3.5","name":"voyage-3.5","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-4-large","name":"voyage-4-large","created_by":"voyage","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-4-lite","name":"voyage-4-lite","created_by":"voyage","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-4","name":"voyage-4","created_by":"voyage","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-code-2","name":"voyage-code-2","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-code-3","name":"voyage-code-3","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.18,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-finance-2","name":"voyage-finance-2","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-law-2","name":"voyage-law-2","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-2-vision","name":"grok-2-vision","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-2","context_window":32800,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":2,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-3-fast","name":"grok-3-fast","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":5,"output":25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-3-mini-fast","name":"grok-3-mini-fast","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.6,"output":4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-3-mini","name":"grok-3-mini","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"batch":true},"pricing":{"input":0.3,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"A lightweight model that thinks before responding.","tagline":"A lightweight model that thinks before responding.","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-3","name":"grok-3","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Grok 3 is the latest model from xAI.","tagline":"Grok 3 is the latest model from xAI.","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4-fast-non-reasoning","name":"grok-4-fast-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-04-17","family":"grok-4","context_window":2000000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4-fast-reasoning","name":"grok-4-fast-reasoning","created_by":"xai","source":"official","last_updated":"2026-04-17","family":"grok-4","context_window":2000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4.1-fast-non-reasoning","name":"grok-4.1-fast-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-17","family":"grok-4.1","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"vision":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.1-fast-reasoning","name":"grok-4.1-fast-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-17","family":"grok-4.1","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-multi-agent-beta","name":"grok-4.20-multi-agent-beta","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":1.25,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-multi-agent","name":"grok-4.20-multi-agent","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"xai/grok-4.20-non-reasoning-beta","name":"grok-4.20-non-reasoning-beta","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":1.25,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-non-reasoning","name":"grok-4.20-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"xai/grok-4.20-reasoning-beta","name":"grok-4.20-reasoning-beta","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-reasoning","name":"grok-4.20-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"xai/grok-4.3","name":"grok-4.3","created_by":"xai","source":"official","family":"grok-4.3","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4","name":"grok-4","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-4","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-build-0.1","name":"grok-build-0.1","created_by":"xai","source":"official","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":1,"output":2},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","license":"proprietary","open_weight":false},{"id":"xai/grok-code-fast-1","name":"grok-code-fast-1","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-code","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":0.2,"output":1.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","tagline":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","max_output_tokens":10000,"license":"proprietary","open_weight":false},{"id":"xai/grok-imagine-image-pro","name":"grok-imagine-image-pro","created_by":"xai","source":"official","last_updated":"2026-03-21","family":"grok-imagine","capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"model_type":"image","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"xai/grok-imagine-image","name":"grok-imagine-image","created_by":"xai","source":"official","last_updated":"2026-03-21","family":"grok-imagine","capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"model_type":"image","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"xai/grok-imagine-video-1.5-preview","name":"grok-imagine-video-1.5-preview","created_by":"xai","source":"official","family":"grok-imagine","model_type":"image","capabilities":{"streaming":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-06-03","license":"proprietary","open_weight":false,"modalities":{"input":["image"],"output":["video"]}},{"id":"xai/grok-imagine-video","name":"grok-imagine-video","created_by":"xai","source":"official","last_updated":"2026-03-21","family":"grok-imagine","capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"model_type":"image","license":"proprietary","open_weight":false,"modalities":{"input":["text","image","video"],"output":["video"]}},{"id":"xiaomi/mimo-v2-flash","name":"mimo-v2-flash","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"fine_tuning":true,"structured_output":true},"pricing":{"input":0.1,"output":0.3},"tools":["function_calling"],"family":"mimo","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"309B MoE reasoning model with 15B active parameters. Fast inference at 150 tokens/s.","tagline":"309B MoE reasoning model with 15B active parameters.","max_output_tokens":64000,"knowledge_cutoff":"2024-12","license":"mit","parameters":309,"active_parameters":15,"architecture":"moe","open_weight":true},{"id":"xiaomi/mimo-v2-pro","name":"mimo-v2-pro","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"structured_output":true},"pricing":{"input":1,"output":3},"tools":["function_calling"],"family":"mimo","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Flagship foundation model with 1M-token context and top-tier agentic capabilities.","tagline":"Flagship foundation model with 1M-token context and top-tier agentic capabilities.","max_output_tokens":128000,"license":"proprietary","parameters":1000,"active_parameters":42,"architecture":"moe","open_weight":false},{"id":"xiaomi/mimo-v2.5-pro","name":"mimo-v2.5-pro","created_by":"xiaomi","source":"official","family":"mimo","context_window":1100000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.43,"output":0.87},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","max_output_tokens":128000},{"id":"xiaomi/mimo-v2.5","name":"mimo-v2.5","created_by":"xiaomi","source":"official","family":"mimo","context_window":1100000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.14,"output":0.28},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","max_output_tokens":128000},{"id":"zai/glm-4.5-air","name":"glm-4.5-air","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":1.1},"family":"glm-4.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.5","name":"glm-4.5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":2.2},"family":"glm-4.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.5v","name":"glm-4.5v","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":66000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":1.8},"family":"glm-4.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.6","name":"glm-4.6","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":2.2},"family":"glm-4.6","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.6v-flash","name":"glm-4.6v-flash","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0,"output":0},"family":"glm-4.6","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.6v","name":"glm-4.6v","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":0.9},"family":"glm-4.6","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.7-flash","name":"glm-4.7-flash","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":200000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.07,"output":0.4},"family":"glm-4.7","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.7-flashx","name":"glm-4.7-flashx","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":200000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.06,"output":0.4},"family":"glm-4.7","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.7","name":"glm-4.7","created_by":"zhipu","source":"official","last_updated":"2026-04-07","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":2.25,"output":2.75},"family":"glm-4.7","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-5-turbo","name":"glm-5-turbo","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.2,"output":4},"family":"glm-5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-5.1","name":"glm-5.1","created_by":"zhipu","source":"official","family":"glm-5.1","context_window":204800,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.3,"output":4.3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"zai/glm-5.2","name":"glm-5.2","created_by":"zhipu","source":"official","family":"glm-5.2","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.4,"output":4.4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"zai/glm-5","name":"glm-5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.8,"output":2.56},"family":"glm-5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-5v-turbo","name":"glm-5v-turbo","created_by":"zhipu","source":"official","family":"glm-5","context_window":200000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.2,"output":4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"}]};
+export const provider: ProviderWithModels = {"id":"vercel","name":"Vercel AI Gateway","region":"US","headquarters":"San Francisco, CA","founded":2015,"url":"https://vercel.com/ai-gateway","api_url":"https://gateway.ai.vercel.com/v1","docs_url":"https://vercel.com/docs/ai-gateway","pricing_url":"https://vercel.com/ai-gateway/models","description":"Vercel AI Gateway provides unified access to 200+ models through the AI SDK.","type":"aggregator","sdk":{"javascript":"ai"},"openai_compatible":true,"free_tier":false,"github_url":"https://github.com/vercel/ai","models_url":"https://sdk.vercel.ai/docs/foundations/providers-and-models","twitter_url":"https://x.com/vercel","blog_url":"https://vercel.com/blog","terms_url":"https://vercel.com/legal/terms","support_url":"https://vercel.com/help","icon":"","models":[{"id":"alibaba/qwen-3-14b","name":"qwen-3-14b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen","context_window":41000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.12,"output":0.24},"model_type":"chat","parameters":14,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen-3-235b","name":"qwen-3-235b","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.09,"output":0.1},"model_type":"chat","parameters":235,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen-3-30b","name":"qwen-3-30b","created_by":"qwen","source":"official","last_updated":"2026-06-10","family":"qwen","context_window":41000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.12,"output":0.5},"model_type":"chat","parameters":30,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen-3-32b","name":"qwen-3-32b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","parameters":32,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen-3.6-max-preview","name":"qwen-3.6-max-preview","created_by":"qwen","source":"official","family":"qwen","context_window":240000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.3,"output":7.8},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-21","description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks.","max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-235b-a22b-thinking","name":"qwen3-235b-a22b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.23,"output":2.3},"model_type":"chat","parameters":235,"active_parameters":22,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-coder-30b-a3b","name":"qwen3-coder-30b-a3b","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","parameters":30,"active_parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-coder-next","name":"qwen3-coder-next","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":1.2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"An open-source code model from Qwen. The latest Qwen3-Coder series delivers strong Coding Agent capabilities. It excels at tool calling and environment interaction. It supports autonomous programming and delivers outstanding coding performance while maintaining broad general-purpose abilities.[How to use](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The open-source Qwen code model.","max_output_tokens":65536,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-coder-plus","name":"qwen3-coder-plus","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1,"output":5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen code models: The latest Qwen3-Coder-Plus series are code generation models built on Qwen3, featuring robust Coding Agent capabilities. They excel at tool calling and environment interaction, enabling autonomous programming with exceptional coding skills while retaining general-purpose functionality. [Usage](https://www.alibabacloud.com/help/en/document_detail/2850166.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://bailian.console.alibabacloud.com/?tab=model#/efm/model_experience_center/text?currentTab=textChat&modelId=qwen3-coder-plus) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Coder is a code generation model.","max_output_tokens":65536,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-coder","name":"qwen3-coder","created_by":"qwen","source":"official","last_updated":"2026-05-07","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.22,"output":1.8},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-embedding-0.6b","name":"qwen3-embedding-0.6b","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","context_window":32800,"model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.01,"output":0},"parameters":0.6},{"id":"alibaba/qwen3-embedding-4b","name":"qwen3-embedding-4b","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","context_window":32800,"model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"parameters":4},{"id":"alibaba/qwen3-embedding-8b","name":"qwen3-embedding-8b","created_by":"qwen","source":"official","last_updated":"2026-03-21","family":"qwen3","context_window":32800,"model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.05,"output":0},"parameters":8},{"id":"alibaba/qwen3-max-preview","name":"qwen3-max-preview","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-max-thinking","name":"qwen3-max-thinking","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3-max","name":"qwen3-max","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Max is the highest-performing model in the Qwen series and excels at complex, multi-step tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt?modelId=qwen3-max-preview) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The most powerful model in the Qwen series, ideal for complex, multi-step tasks.","max_output_tokens":32768,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-next-80b-a3b-instruct","name":"qwen3-next-80b-a3b-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.15,"output":1.2},"model_type":"chat","parameters":80,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-next-80b-a3b-thinking","name":"qwen3-next-80b-a3b-thinking","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.15,"output":1.2},"model_type":"chat","parameters":80,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-vl-235b-a22b-instruct","name":"qwen3-vl-235b-a22b-instruct","created_by":"qwen","source":"official","family":"qwen3","context_window":262100,"model_type":"chat","parameters":235,"active_parameters":22,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.88},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","description":"Alibaba Cloud's open-source Qwen-VL edition. [Usage](https://www.alibabacloud.com/help/en/document_detail/2845871.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html#15ef0a40798a3) Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The Qwen-VL open-source model from Alibaba Cloud.","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3-vl-instruct","name":"qwen3-vl-instruct","created_by":"qwen","source":"official","last_updated":"2026-06-29","family":"qwen3","context_window":262100,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.2,"output":0.88},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"alibaba/qwen3-vl-thinking","name":"qwen3-vl-thinking","created_by":"qwen","source":"official","last_updated":"2026-04-07","family":"qwen3","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.4,"output":3.95},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"alibaba/qwen3.5-flash","name":"qwen3.5-flash","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Flash is the fastest and most cost-effective model in the Qwen series, designed for simple jobs. Qwen-Flash features flexible tiered pricing, resulting in more reasonable billing compared to Qwen-Turbo. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"The fastest and most cost-effective model in the Qwen series, ideal for simple tasks.","max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.5-plus","name":"qwen3.5-plus","created_by":"qwen","source":"official","last_updated":"2026-03-23","family":"qwen3.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.4,"output":2.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"A balanced model with inference performance, cost, and speed between Qwen-Max and Qwen-Flash, ideal for moderately complex tasks.","max_output_tokens":64000,"license":"proprietary","open_weight":false},{"id":"alibaba/qwen3.6-27b","name":"qwen3.6-27b","created_by":"qwen","source":"official","family":"qwen3.6","context_window":256000,"model_type":"chat","parameters":27,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":3.6},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.6-plus","name":"qwen3.6-plus","created_by":"qwen","source":"official","family":"qwen3.6","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03","description":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks. [Usage](https://www.alibabacloud.com/help/en/document_detail/2841718.html#24e54b27d4agt) \\| [Thinking](https://www.alibabacloud.com/help/en/document_detail/2870973.html) \\| [API reference](https://www.alibabacloud.com/help/en/document_detail/2712576.html) \\| [Try online](https://modelstudio.console.alibabacloud.com/?tab=playground#/efm/prompt) > Qwen3.5 Plus supports text, image, and video inputs. Its performance on plain text tasks is comparable to that of Qwen3 Max, while offering superior efficiency at a lower cost. Its multimodal capabilities are a significant improvement over the Qwen3 VL series. #### International If you select the [International](https://www.alibabacloud.com/help/en/document_detail/3004398.html#080da663a75xh) deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.","tagline":"Qwen-Plus offers balanced capabilities: inference quality, cost, and speed are between Qwen-Max and Qwen-Flash, making it ideal for medium-complexity tasks.","max_output_tokens":64000,"license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.7-max","name":"qwen3.7-max","created_by":"qwen","source":"official","family":"qwen3.7","context_window":991000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":1.25,"output":3.75},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02","license":"apache-2.0","open_weight":true},{"id":"alibaba/qwen3.7-plus","name":"qwen3.7-plus","created_by":"qwen","source":"official","family":"qwen3.7","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"fine_tuning":true,"json_mode":true},"pricing":{"input":0.32,"output":1.28},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18","license":"apache-2.0","open_weight":true},{"id":"alibaba/wan-v2.5-t2v-preview","name":"wan-v2.5-t2v-preview","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-i2v-flash","name":"wan-v2.6-i2v-flash","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-i2v","name":"wan-v2.6-i2v","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-r2v-flash","name":"wan-v2.6-r2v-flash","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-r2v","name":"wan-v2.6-r2v","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"alibaba/wan-v2.6-t2v","name":"wan-v2.6-t2v","created_by":"qwen","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"family":"wan","model_type":"video"},{"id":"amazon/nova-2-lite","name":"nova-2-lite","created_by":"amazon","source":"official","last_updated":"2026-06-29","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"amazon/nova-lite","name":"nova-lite","created_by":"amazon","source":"official","last_updated":"2026-06-29","context_window":300000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.06,"output":0.24},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"amazon/nova-micro","name":"nova-micro","created_by":"amazon","source":"official","last_updated":"2026-06-29","context_window":128000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.04,"output":0.14},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"amazon/nova-pro","name":"nova-pro","created_by":"amazon","source":"official","last_updated":"2026-06-29","context_window":300000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.8,"output":3.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"amazon/titan-embed-text-v2","name":"titan-embed-text-v2","created_by":"amazon","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0}},{"id":"anthropic/claude-3-haiku","name":"claude-3-haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true},"pricing":{"input":0.25,"output":1.25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"proprietary","open_weight":false},{"id":"anthropic/claude-3-opus","name":"claude-3-opus","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"batch":true},"pricing":{"input":15,"output":75},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"license":"proprietary","open_weight":false},{"id":"anthropic/claude-3.5-haiku","name":"claude-3.5-haiku","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.8,"output":4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-3.5-sonnet-20240620","name":"claude-3.5-sonnet-20240620","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":200000,"alias":"anthropic/claude-3.5-sonnet","capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-3.5-sonnet","name":"claude-3.5-sonnet-20240620","created_by":"anthropic","source":"official","last_updated":"2026-03-26","family":"claude-sonnet","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":3,"output":15},"snapshots":["anthropic/claude-3.5-sonnet-20240620"],"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-3.7-sonnet","name":"claude-3.7-sonnet","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-fable-5","name":"claude-fable-5","created_by":"anthropic","source":"official","family":"claude","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":10,"output":50},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10","description":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","tagline":"Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work","max_output_tokens":128000,"license":"proprietary","open_weight":false},{"id":"anthropic/claude-haiku-4.5","name":"claude-haiku-4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1,"output":5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.1","name":"claude-opus-4.1","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":15,"output":75},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.5","name":"claude-opus-4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.6","name":"claude-opus-4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-opus-4.7","name":"claude-opus-4.7","created_by":"anthropic","source":"official","family":"claude-opus","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-17"},{"id":"anthropic/claude-opus-4.8","name":"claude-opus-4.8","created_by":"anthropic","source":"official","family":"claude-opus","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":25},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29"},{"id":"anthropic/claude-opus-4","name":"claude-opus-4","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":15,"output":75},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-sonnet-4.5","name":"claude-sonnet-4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-sonnet-4.6","name":"claude-sonnet-4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"anthropic/claude-sonnet-4","name":"claude-sonnet-4","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"arcee-ai/trinity-large-preview","name":"trinity-large-preview","created_by":"arcee","source":"official","last_updated":"2026-03-23","context_window":131000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.25,"output":1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"arcee-ai/trinity-large-thinking","name":"trinity-large-thinking","created_by":"arcee","source":"official","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.25,"output":0.9},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"arcee-ai/trinity-mini","name":"trinity-mini","created_by":"arcee","source":"official","last_updated":"2026-06-29","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.04,"output":0.15},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"bfl/flux-2-flex","name":"flux-2-flex","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"FLUX.2 variant with adjustable steps and guidance, tuned for typography and editing workflows. Up to 8 references via API.","tagline":"Flexible FLUX.2 for typography and editing.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-klein-4b","name":"flux-2-klein-4b","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"fine_tuning":true},"pricing":{"input":0,"output":0},"model_type":"image","parameters":4,"description":"Lightweight 4B-parameter FLUX.2 flow model for sub-second inference; ~13 GB VRAM. Open base weights under Apache 2.0.","tagline":"Fastest, most lightweight FLUX.2.","license":"apache-2.0","open_weight":true,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-klein-9b","name":"flux-2-klein-9b","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true,"fine_tuning":true},"pricing":{"input":0,"output":0},"model_type":"image","parameters":9,"description":"Balanced 9B FLUX.2 flow model with 8B text embedder; sub-second inference at ~24 GB VRAM. Open base weights under FLUX non-commercial license.","tagline":"Balanced quality and speed for [klein].","license":"flux-1-dev-non-commercial","open_weight":true,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-max","name":"flux-2-max","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","context_window":67300,"capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Highest-quality FLUX.2 model with web grounding search, up to 8 references via API and 10 in playground.","tagline":"Highest quality FLUX.2.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-2-pro","name":"flux-2-pro","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","context_window":67300,"capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Production-grade default model for FLUX.2 text-to-image generation and image editing. Supports up to 10 reference images and 4MP output.","tagline":"Recommended default for image generation and editing.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-kontext-max","name":"flux-kontext-max","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Higher-quality variant of Kontext with stronger editing fidelity. Legacy; superseded by FLUX.2 [max] for new projects.","tagline":"Legacy FLUX.1 Kontext [max].","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-kontext-pro","name":"flux-kontext-pro","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Context-aware image editing and text-to-image. Legacy FLUX.1 line; BFL recommends FLUX.2 [pro] for new projects.","tagline":"Legacy FLUX.1 Kontext [pro].","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-pro-1.0-fill","name":"flux-pro-1.0-fill","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Text-driven inpainting and outpainting using an input image and mask.","tagline":"Inpainting and outpainting for FLUX.1.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-pro-1.1-ultra","name":"flux-pro-1.1-ultra","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"FLUX 1.1 [pro] with up to 4MP resolution and optional Raw mode for photographic authenticity. Supports image-to-image.","tagline":"FLUX 1.1 [pro] up to 4MP with raw mode.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"bfl/flux-pro-1.1","name":"flux-pro-1.1","created_by":"black-forest-labs","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Fast and reliable text-to-image baseline with strong prompt adherence. Sometimes called FLUX1.1 [pro].","tagline":"Fast and reliable FLUX 1.1 baseline.","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seed-1.6","name":"seed-1.6","created_by":"bytedance","source":"official","last_updated":"2026-06-29","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":0.25,"output":2},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Seed1.6 integrates multimodal capabilities, with \"adaptive thinking\" balancing model performance with reasoning capability.","tagline":"Seed1.6 integrates multimodal capabilities, with \"adaptive thinking\" balancing model performance with reasoning capability."},{"id":"bytedance/seed-1.8","name":"seed-1.8","created_by":"bytedance","source":"official","last_updated":"2026-06-29","context_window":256000,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true},"pricing":{"input":0.25,"output":2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"description":"A general-purpose agent model capable of efficiently and accurately completing tasks in complex workflows.","tagline":"A general-purpose agent model capable of efficiently and accurately completing tasks in complex workflows."},{"id":"bytedance/seedance-2.0-fast","name":"seedance-2.0-fast","created_by":"bytedance","source":"official","model_type":"video","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16"},{"id":"bytedance/seedance-2.0","name":"seedance-2.0","created_by":"bytedance","source":"official","model_type":"video","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16","description":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","tagline":"Adopts a unified multimodal audio-video joint generation architecture that supports text, image, audio, and video inputs, leading to the most comprehensive multimodal content reference and editing cap","modalities":{"input":["text","audio"],"output":["video"]}},{"id":"bytedance/seedance-v1.0-lite-i2v","name":"seedance-v1.0-lite-i2v","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.0-lite-t2v","name":"seedance-v1.0-lite-t2v","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.0-pro-fast","name":"seedance-v1.0-pro-fast","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.0-pro","name":"seedance-v1.0-pro","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedance-v1.5-pro","name":"seedance-v1.5-pro","created_by":"bytedance","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"bytedance/seedream-4.0","name":"seedream-4.0","created_by":"bytedance","source":"official","model_type":"image","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-12","description":"A new-generation image creation model can flexibly handle complex multimodal tasks, producing high-definition images at up to 4K resolution.","tagline":"A new-generation image creation model can flexibly handle complex multimodal tasks, producing high-definition images at up to 4K resolution.","modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seedream-4.5","name":"seedream-4.5","created_by":"bytedance","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-12","description":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","tagline":"Seedream 4.5 has an all-round improvement to support professional visual creatives with high consistency and fidelity.","modalities":{"input":["text"],"output":["image"]}},{"id":"bytedance/seedream-5.0-lite","name":"seedream-5.0-lite","created_by":"bytedance","source":"official","model_type":"image","capabilities":{"streaming":true,"vision":true,"reasoning":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-12","description":"A unified multimodal image generation model endowed with deep thinking and online search capabilities.","tagline":"A unified multimodal image generation model endowed with deep thinking and online search capabilities.","modalities":{"input":["text"],"output":["image"]}},{"id":"cohere/command-a","name":"command-a","created_by":"cohere","source":"official","last_updated":"2026-03-23","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":2.5,"output":10},"family":"command-a","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"cohere/embed-v4.0","name":"embed-v4.0","created_by":"cohere","source":"official","last_updated":"2026-06-18","model_type":"embed","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.12,"output":0},"family":"embed","context_window":128000,"description":"A model that allows for text and images to be classified or turned into embeddings","tagline":"Text, Images, Mixed texts/images (i.e.","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"cohere/rerank-v3.5","name":"rerank-v3.5","created_by":"cohere","source":"official","family":"rerank","context_window":4100,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-14","description":"A model that allows for re-ranking English Language documents and semi-structured data (JSON). This model has a context length of 4096 tokens.","tagline":"Unique per deployment","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"cohere/rerank-v4-fast","name":"rerank-v4-fast","created_by":"cohere","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16"},{"id":"cohere/rerank-v4-pro","name":"rerank-v4-pro","created_by":"cohere","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-04-16"},{"id":"deepseek/deepseek-r1","name":"deepseek-r1","created_by":"deepseek","source":"official","last_updated":"2026-06-29","family":"deepseek-r1","context_window":160000,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true},"pricing":{"input":1.35,"output":5.4},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"deepseek/deepseek-v3.1-terminus","name":"deepseek-v3.1-terminus","created_by":"deepseek","source":"official","last_updated":"2026-03-23","family":"deepseek","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.27,"output":1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.1","name":"deepseek-v3.1","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.21,"output":0.79},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.2-thinking","name":"deepseek-v3.2-thinking","created_by":"deepseek","source":"official","last_updated":"2026-06-29","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":0.26,"output":0.38},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3.2","name":"deepseek-v3.2","created_by":"deepseek","source":"official","last_updated":"2026-05-01","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"reasoning":true},"pricing":{"input":0.28,"output":0.42},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v3","name":"deepseek-v3","created_by":"deepseek","source":"official","last_updated":"2026-06-02","family":"deepseek","context_window":163800,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.27,"output":1.12},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"deepseek/deepseek-v4-flash","name":"deepseek-v4-flash","created_by":"deepseek","source":"official","family":"deepseek","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"fine_tuning":true},"pricing":{"input":0.14,"output":0.28},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","description":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Flash hybrid model with both non-thinking and thinking (default) modes.","max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"deepseek/deepseek-v4-pro","name":"deepseek-v4-pro","created_by":"deepseek","source":"official","family":"deepseek","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"fine_tuning":true},"pricing":{"input":0.43,"output":0.87},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","description":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","tagline":"DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.","max_output_tokens":384000,"license":"mit","open_weight":true},{"id":"google/gemini-2.0-flash-lite","name":"gemini-2.0-flash-lite","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"json_mode":true},"pricing":{"input":0.07,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":8192,"knowledge_cutoff":"2024-08","license":"proprietary","open_weight":false},{"id":"google/gemini-2.0-flash","name":"gemini-2.0-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"json_mode":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":8192,"knowledge_cutoff":"2024-08","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-image","name":"gemini-2.5-flash-image","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":32800,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Image (Nano Banana) Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency. Gemini 2.5 Flash Image, also known as Nano Banana, is best for high-volume generation, conversational image editing, and low-latency creative workflows that require native multimodal understanding.","tagline":"Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency.","max_output_tokens":32768,"knowledge_cutoff":"2025-06","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite-preview-09-2025","name":"gemini-2.5-flash-lite-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-lite","name":"gemini-2.5-flash-lite","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash-Lite Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 2.5 Flash-Lite is best for high-volume classification, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash-preview-09-2025","name":"gemini-2.5-flash-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-flash","name":"gemini-2.5-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.3,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Our best model in terms of price-performance, offering well-rounded capabilities. 2.5 Flash is best for large scale processing, low-latency, high volume tasks that require thinking, and agentic use cases.","tagline":"Our best model in terms of price-performance, offering well-rounded capabilities.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-2.5-pro","name":"gemini-2.5-pro","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Pro Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","tagline":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-flash","name":"gemini-3-flash","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemini-3-pro-image","name":"gemini-3-pro-image","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","context_window":65500,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3 Pro Image Nano Banana Pro is a sophisticated reasoning-driven engine for professional-grade image editing and generation, offering studio-quality precision and advanced creative control. Nano Banana Pro is best for complex graphic design, high-fidelity product mockups, and factual data visualizations that require accurate text rendering and real-world grounding via Google Search.","tagline":"The Interactions API is now generally available.","max_output_tokens":32768,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3-pro-preview","name":"gemini-3-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image-preview","name":"gemini-3.1-flash-image-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","context_window":131100,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"max_output_tokens":32768,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-image","name":"gemini-3.1-flash-image","created_by":"google","source":"official","family":"gemini-3.1","context_window":131100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.5,"output":3},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-10","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash Image Nano Banana 2 provides high-quality image generation and conversational editing at a mainstream price point and low latency. It serves as the high-efficiency counterpart to Gemini 3 Pro Image , optimized for speed and high-volume developer use cases. Key updates: New output resolution options: New support for 0.5K, 2K and 4K, default 1K New Image Search Grounding: Integration of both text and image search results to inform generation with real-time web data Supported with Thinking on or off New 1:4, 4:1, 1:8 and 8:1 aspect ratios Improved aspect ratio adherence Improved image quality and consistency Improved i18n text rendering","tagline":"The Interactions API is now generally available.","max_output_tokens":32768,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite-preview","name":"gemini-3.1-flash-lite-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 3.1 Flash-Lite is best for high-volume agentic tasks, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-flash-lite","name":"gemini-3.1-flash-lite","created_by":"google","source":"official","family":"gemini-3.1","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":0.25,"output":1.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-10","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash-Lite Gemini 3.1 Flash-Lite is a low-latency, cost-effective multimodal model optimized for high-frequency, lightweight tasks. The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary constraints.","tagline":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary con","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.1-pro-preview","name":"gemini-3.1-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-3.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":2,"output":12},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Pro Preview Built to refine the performance and reliability of the Gemini 3 Pro series, Gemini 3.1 Pro Preview provides better thinking, improved token efficiency, and a more grounded, factually consistent experience. It's optimized for software engineering behavior and usability, as well as agentic workflows requiring precise tool usage and reliable multi-step execution across real-world domains.","tagline":"The Interactions API is now generally available.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-3.5-flash","name":"gemini-3.5-flash","created_by":"google","source":"official","family":"gemini-3.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"batch":true,"json_mode":true},"pricing":{"input":1.5,"output":9},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.5 Flash Gemini 3.5 Flash provides sustained frontier-level intelligence optimized for real-world tasks at a higher speed and lower cost. Designed for the agentic era, it excels at sub-agent deployment, multi-step workflows, and long-horizon tasks at scale. This model is particularly effective for rapid agentic loops involving complex coding cycles and iterations.","tagline":"The Interactions API is now generally available.","max_output_tokens":65536,"knowledge_cutoff":"2025-01","license":"proprietary","open_weight":false},{"id":"google/gemini-embedding-001","name":"gemini-embedding-001","created_by":"google","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.15,"output":0},"family":"gemini-embedding","context_window":2048,"license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemini-embedding-2","name":"gemini-embedding-2","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-embedding","model_type":"embed","capabilities":{"streaming":true,"vision":true},"pricing":{"input":0.2,"output":0},"context_window":8192,"license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"google/gemma-4-26b-a4b-it","name":"gemma-4-26b-a4b-it","created_by":"google","source":"official","family":"gemma-4","context_window":262100,"model_type":"chat","parameters":26,"active_parameters":4,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.13,"output":0.4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"google/gemma-4-31b-it","name":"gemma-4-31b-it","created_by":"google","source":"official","family":"gemma-4","context_window":262100,"model_type":"chat","parameters":31,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.14,"output":0.4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"},{"id":"google/imagen-4.0-fast-generate-001","name":"imagen-4.0-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","family":"imagen-4.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"google/imagen-4.0-generate-001","name":"imagen-4.0-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","family":"imagen-4.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"google/imagen-4.0-ultra-generate-001","name":"imagen-4.0-ultra-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","family":"imagen-4.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"google/text-embedding-005","name":"text-embedding-005","created_by":"google","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0},"family":"text-embedding"},{"id":"google/text-multilingual-embedding-002","name":"text-multilingual-embedding-002","created_by":"google","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.03,"output":0}},{"id":"google/veo-3.0-fast-generate-001","name":"veo-3.0-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]}},{"id":"google/veo-3.0-generate-001","name":"veo-3.0-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.0","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]}},{"id":"google/veo-3.1-fast-generate-001","name":"veo-3.1-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.1"},{"id":"google/veo-3.1-generate-001","name":"veo-3.1-generate-001","created_by":"google","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video","family":"veo-3.1"},{"id":"inception/mercury-2","name":"mercury-2","created_by":"inception","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.25,"output":0.75},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Mercury is the first diffusion large language model (dLLM). Applying a breakthrough discrete diffusion approach, the model runs 5-10x faster than even speed optimized models like Claude 3.5 Haiku and GPT-4o Mini while matching their performance. Mercury's speed means that users can stay in the flow, enjoying rapid chat-based iteration and responsive completion suggestions. On Copilot Arena, Mercury ranks 1st in speed and ties for 2nd in quality.","tagline":"The fastest reasoning dLLM.","max_output_tokens":50000,"license":"proprietary","open_weight":false},{"id":"inception/mercury-coder-small","name":"mercury-coder-small","created_by":"inception","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.25,"output":1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"interfaze/interfaze-beta","name":"interfaze-beta","created_by":"interfaze","source":"official","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"reasoning":true,"vision":true,"tool_call":true},"pricing":{"input":1.5,"output":3.5},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29","tools":["function_calling"]},{"id":"klingai/kling-v2.5-turbo-i2v","name":"kling-v2.5-turbo-i2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.5-turbo-t2v","name":"kling-v2.5-turbo-t2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.6-i2v","name":"kling-v2.6-i2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.6-motion-control","name":"kling-v2.6-motion-control","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v2.6-t2v","name":"kling-v2.6-t2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v3.0-i2v","name":"kling-v3.0-i2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"klingai/kling-v3.0-motion-control","name":"kling-v3.0-motion-control","created_by":"klingai","source":"official","model_type":"video","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"klingai/kling-v3.0-t2v","name":"kling-v3.0-t2v","created_by":"klingai","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"video"},{"id":"kwaipilot/kat-coder-pro-v1","name":"kat-coder-pro-v1","created_by":"kwaipilot","source":"official","last_updated":"2026-06-29","context_window":256000,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true},"pricing":{"input":0.3,"output":1.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"kwaipilot/kat-coder-pro-v2","name":"kat-coder-pro-v2","created_by":"kwaipilot","source":"official","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-01"},{"id":"meituan/longcat-flash-chat","name":"longcat-flash-chat","created_by":"meituan","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0,"output":0},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"meituan/longcat-flash-thinking-2601","name":"longcat-flash-thinking-2601","created_by":"meituan","source":"official","last_updated":"2026-06-29","context_window":32800,"alias":"meituan/longcat-flash-thinking","capabilities":{"streaming":true,"reasoning":true,"tool_call":true},"pricing":{"input":0,"output":0},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"meituan/longcat-flash-thinking","name":"longcat-flash-thinking-2601","created_by":"meituan","source":"official","last_updated":"2026-03-26","context_window":32800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0,"output":0},"snapshots":["meituan/longcat-flash-thinking-2601"],"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.1-70b","name":"llama-3.1-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.1","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.72,"output":0.72},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.1-8b","name":"llama-3.1-8b","created_by":"meta","source":"official","last_updated":"2026-06-02","family":"llama-3.1","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.02,"output":0.05},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-11b","name":"llama-3.2-11b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.16,"output":0.16},"model_type":"chat","parameters":11,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-1b","name":"llama-3.2-1b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","parameters":1,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-3b","name":"llama-3.2-3b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","parameters":3,"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.2-90b","name":"llama-3.2-90b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.2","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.72,"output":0.72},"model_type":"chat","parameters":90,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-3.3-70b","name":"llama-3.3-70b","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-3.3","context_window":128000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.59,"output":0.72},"model_type":"chat","parameters":70,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-4-maverick","name":"llama-4-maverick","created_by":"meta","source":"official","last_updated":"2026-04-07","family":"llama-4","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.24,"output":0.97},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"meta/llama-4-scout","name":"llama-4-scout","created_by":"meta","source":"official","last_updated":"2026-03-23","family":"llama-4","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.17,"output":0.66},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"minimax/minimax-m2.1-lightning","name":"minimax-m2.1-lightning","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":2.4},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"minimax/minimax-m2.1","name":"minimax-m2.1","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.3,"output":1.2},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","tagline":"Powerful Multi-Language Programming Capabilities with Comprehensively Enhanced Programming Experience","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5-highspeed","name":"minimax-m2.5-highspeed","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":2.4},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"M2.5 highspeed: Same performance, faster and more agile","tagline":"M2.5 highspeed: Same performance, faster and more agile","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.5","name":"minimax-m2.5","created_by":"minimax","source":"official","last_updated":"2026-06-02","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.07,"output":0.57},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Peak Performance. Ultimate Value. Master the Complex","tagline":"Peak Performance.","license":"mit","open_weight":true},{"id":"minimax/minimax-m2.7-highspeed","name":"minimax-m2.7-highspeed","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":2.4},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"M2.7 Highspeed: Same performance, faster and more agile","tagline":"M2.7 Highspeed: Same performance, faster and more agile","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2.7","name":"minimax-m2.7","created_by":"minimax","source":"official","last_updated":"2026-06-02","context_window":204800,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.15,"output":0.6},"family":"minimax","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Beginning the journey of recursive self-improvement","tagline":"Beginning the journey of recursive self-improvement","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m2","name":"minimax-m2","created_by":"minimax","source":"official","last_updated":"2026-03-23","context_window":205000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"model_type":"chat","family":"minimax","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Agentic capabilities, Advanced reasoning","tagline":"Agentic capabilities, Advanced reasoning","license":"proprietary","open_weight":false},{"id":"minimax/minimax-m3","name":"minimax-m3","created_by":"minimax","source":"official","family":"minimax","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":1.2},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02","description":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","tagline":"Latest M-series language model for agentic reasoning, tool use, coding, and long-context tasks","license":"proprietary","open_weight":false},{"id":"mistral/codestral-embed","name":"codestral-embed","created_by":"mistral","source":"official","last_updated":"2026-03-21","family":"codestral","model_type":"embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0},"context_window":8000,"license":"mnpl","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]}},{"id":"mistral/codestral","name":"codestral","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"codestral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.3,"output":0.9},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our first cutting-edge language model for coding and FIM, released under the MNPL license.","tagline":"Our cutting-edge language model for code completion released end of July 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM) and code generation.","license":"mnpl","open_weight":true},{"id":"mistral/devstral-2","name":"devstral-2","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.4,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/devstral-small-2","name":"devstral-small-2","created_by":"mistral","source":"official","last_updated":"2026-06-29","family":"devstral","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"vision":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/devstral-small","name":"devstral-small","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"devstral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"A 24B text model, open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","tagline":"Our open source model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.","license":"apache-2.0","open_weight":true},{"id":"mistral/magistral-medium","name":"magistral-medium","created_by":"mistral","source":"official","last_updated":"2026-06-29","context_window":128000,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true},"pricing":{"input":2,"output":5},"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"description":"Our first frontier-class reasoning model released June 2025.","tagline":"Our frontier-class multimodal reasoning model update of September 2025.","license":"proprietary","open_weight":false},{"id":"mistral/magistral-small","name":"magistral-small","created_by":"mistral","source":"official","last_updated":"2026-06-29","context_window":128000,"capabilities":{"streaming":true,"vision":true,"reasoning":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","family":"magistral","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"description":"Our small reasoning model released July 2025.","tagline":"Our small multimodal reasoning model update of September 2025.","license":"apache-2.0","open_weight":true},{"id":"mistral/ministral-14b","name":"ministral-14b","created_by":"mistral","source":"official","last_updated":"2026-06-29","family":"ministral","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.2,"output":0.2},"model_type":"chat","parameters":14,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"license":"apache-2.0","open_weight":true},{"id":"mistral/ministral-3b","name":"ministral-3b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.1},"model_type":"chat","parameters":3,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"mistral/ministral-8b","name":"ministral-8b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"ministral","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","parameters":8,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"apache-2.0","open_weight":true},{"id":"mistral/mistral-embed","name":"mistral-embed","created_by":"mistral","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0},"family":"mistral-embed","context_window":8000,"license":"apache-2.0","open_weight":true,"modalities":{"input":["text","image"],"output":["text"]}},{"id":"mistral/mistral-large-3","name":"mistral-large-3","created_by":"mistral","source":"official","last_updated":"2026-06-29","family":"mistral-large","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"mistral/mistral-medium-3.5","name":"mistral-medium-3.5","created_by":"mistral","source":"official","family":"mistral-medium","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":1.5,"output":7.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"mistral/mistral-medium","name":"mistral-medium","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-medium","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0.4,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our first SOTA enterprise grade model.","tagline":"Our frontier-class multimodal model released August 2025.","license":"proprietary","open_weight":false},{"id":"mistral/mistral-nemo","name":"mistral-nemo","created_by":"mistral","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","family":"mistral-nemo","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Our best multilingual open source model released July 2024.","tagline":"Our best multilingual open source model released July 2024.","license":"apache-2.0","open_weight":true},{"id":"mistral/mistral-small","name":"mistral-small","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-small","context_window":32000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.1,"output":0.3},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"An enterprise grade small model for efficiency.","tagline":"Our powerful hybrid model unifying instruct, reasoning, and coding capabilities in a single model.","license":"apache-2.0","open_weight":true},{"id":"mistral/mixtral-8x22b-instruct","name":"mixtral-8x22b-instruct","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mixtral","context_window":65500,"capabilities":{"streaming":true},"pricing":{"input":1.2,"output":1.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"mistral/pixtral-12b","name":"pixtral-12b","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"pixtral","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":0.15,"output":0.15},"model_type":"chat","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"A 12B model with image understanding capabilities in addition to text.","tagline":"A 12B model with image understanding capabilities in addition to text.","license":"apache-2.0","open_weight":true},{"id":"mistral/pixtral-large","name":"pixtral-large","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"pixtral","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":true},"pricing":{"input":2,"output":6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"license":"mnpl","open_weight":true},{"id":"moonshotai/kimi-k2-0905","name":"kimi-k2-0905","created_by":"moonshot","source":"official","last_updated":"2026-04-16","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.6,"output":2.5},"family":"kimi-k2","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"moonshotai/kimi-k2-thinking-turbo","name":"kimi-k2-thinking-turbo","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":1.15,"output":8},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Deep reasoning with high speed output.","tagline":"Deep reasoning with high speed output.","max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2-thinking","name":"kimi-k2-thinking","created_by":"moonshot","source":"official","last_updated":"2026-06-29","context_window":216100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.47,"output":2},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Long-term thinking, multi-step tool usage, complex problem solving.","tagline":"Long-term thinking, multi-step tool usage, complex problem solving.","max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2-turbo","name":"kimi-k2-turbo","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":256000,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":1.15,"output":8},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"moonshotai/kimi-k2.5","name":"kimi-k2.5","created_by":"moonshot","source":"official","last_updated":"2026-06-18","context_window":262100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"fine_tuning":true},"pricing":{"input":0.6,"output":3},"family":"kimi-k2.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","tagline":"Kimi's most intelligent model with native multimodal support, thinking/non-thinking modes.","max_output_tokens":32768,"license":"mit","open_weight":true},{"id":"moonshotai/kimi-k2.6","name":"kimi-k2.6","created_by":"moonshot","source":"official","family":"kimi-k2.6","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.95,"output":4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"moonshotai/kimi-k2.7-code-highspeed","name":"kimi-k2.7-code-highspeed","created_by":"moonshot","source":"official","family":"kimi-k2.7","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.9,"output":8},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"moonshotai/kimi-k2.7-code","name":"kimi-k2.7-code","created_by":"moonshot","source":"official","family":"kimi-k2.7","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.74,"output":3.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-18"},{"id":"moonshotai/kimi-k2","name":"kimi-k2","created_by":"moonshot","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.57,"output":2.3},"family":"kimi-k2","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"morph/morph-v3-fast","name":"morph-v3-fast","created_by":"morph","source":"official","last_updated":"2026-03-23","context_window":81900,"capabilities":{"streaming":true},"pricing":{"input":0.8,"output":1.2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"morph/morph-v3-large","name":"morph-v3-large","created_by":"morph","source":"official","last_updated":"2026-03-23","context_window":81900,"capabilities":{"streaming":true},"pricing":{"input":0.9,"output":1.9},"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-3-nano-30b-a3b","name":"nemotron-3-nano-30b-a3b","created_by":"nvidia","source":"official","last_updated":"2026-06-29","context_window":262100,"capabilities":{"streaming":true,"reasoning":true,"tool_call":true},"pricing":{"input":0.05,"output":0.24},"model_type":"chat","parameters":30,"active_parameters":3,"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"nvidia/nemotron-3-super-120b-a12b","name":"nemotron-3-super-120b-a12b","created_by":"nvidia","source":"official","context_window":256000,"model_type":"chat","parameters":120,"active_parameters":12,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.15,"output":0.65},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29","tools":["function_calling"]},{"id":"nvidia/nemotron-3-ultra-550b-a55b","name":"nemotron-3-ultra-550b-a55b","created_by":"nvidia","source":"official","context_window":1000000,"model_type":"chat","parameters":550,"active_parameters":55,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.37,"output":1.08},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-10"},{"id":"nvidia/nemotron-nano-12b-v2-vl","name":"nemotron-nano-12b-v2-vl","created_by":"nvidia","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":0.6},"model_type":"chat","parameters":12,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"nvidia/nemotron-nano-9b-v2","name":"nemotron-nano-9b-v2","created_by":"nvidia","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.06,"output":0.23},"model_type":"chat","parameters":9,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-3.5-turbo-instruct","name":"gpt-3.5-turbo-instruct","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-3.5","context_window":8200,"capabilities":{"streaming":true,"fine_tuning":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"reasoning":false},"pricing":{"input":1.5,"output":2},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Similar capabilities as GPT-3 era models. Compatible with legacy Completions endpoint and not Chat Completions.","tagline":"An older model only compatible with the legacy Completions endpoint","max_output_tokens":4096,"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-3.5-turbo","name":"gpt-3.5-turbo","created_by":"openai","source":"official","last_updated":"2026-06-29","family":"gpt-3.5","context_window":16400,"capabilities":{"streaming":true,"tool_call":true,"fine_tuning":true,"vision":false,"structured_output":false,"json_mode":false,"reasoning":false,"batch":true},"pricing":{"input":0.5,"output":1.5},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"description":"GPT-3.5 Turbo models can understand and generate natural language or code and have been optimized for chat using the Chat Completions API but work well for non-chat tasks as well. As of July 2024, use gpt-4o-mini in place of GPT-3.5 Turbo, as it is cheaper, more capable, multimodal, and just as fast. GPT-3.5 Turbo is still available for use in the API.","tagline":"Legacy GPT model for cheaper chat and non-chat tasks","max_output_tokens":4096,"knowledge_cutoff":"2021-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-4-turbo","name":"gpt-4-turbo","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4-turbo","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":10,"output":30},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4 Turbo is the next generation of GPT-4, an older high-intelligence GPT model. It was designed to be a cheaper, better version of GPT-4. Today, we recommend using a newer model like GPT-4o.","tagline":"An older high-intelligence GPT model","max_output_tokens":4096,"knowledge_cutoff":"2023-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-mini","name":"gpt-4.1-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":0.4,"output":1.6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4.1 mini excels at instruction following and tool calling. It features a","tagline":"Smaller, faster version of GPT-4.1","max_output_tokens":32768,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1-nano","name":"gpt-4.1-nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":0.1,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4.1 nano excels at instruction following and tool calling. It features a","tagline":"Fastest, most cost-efficient version of GPT-4.1","max_output_tokens":32768,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4.1","name":"gpt-4.1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4.1","context_window":1000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":2,"output":8},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4.1 excels at instruction following and tool calling, with broad","tagline":"Smartest non-reasoning model","max_output_tokens":32768,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini-search-preview","name":"gpt-4o-mini-search-preview","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4o","context_window":128000,"capabilities":{"streaming":true,"vision":true,"structured_output":true,"json_mode":true,"tool_call":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"GPT-4o mini Search Preview is a specialized model trained to understand and execute [web search](/api/docs/guides/tools-web-search?api-mode=chat) queries with the Chat Completions API. In addition to token fees, web search queries have a fee per tool call. Learn more in the [pricing](/api/docs/pricing) page.","tagline":"Fast, affordable small model for web search","max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-mini-transcribe","name":"gpt-4o-mini-transcribe","created_by":"openai","source":"official","family":"gpt-4o","model_type":"transcription","capabilities":{"streaming":true},"pricing":{"input":1.25,"output":5},"last_updated":"2026-06-29","description":"GPT-4o mini Transcribe is a speech-to-text model that uses GPT-4o mini to transcribe audio.","tagline":"Speech-to-text model powered by GPT-4o mini","context_window":16000,"max_output_tokens":2000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false,"modalities":{"input":["audio","text"],"output":["text"]}},{"id":"openai/gpt-4o-mini","name":"gpt-4o-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4o","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":0.15,"output":0.6},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4o mini (“o” for “omni”) is a fast, affordable small model for focused tasks.","tagline":"Fast, affordable small model for focused tasks","max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-4o-transcribe","name":"gpt-4o-transcribe","created_by":"openai","source":"official","family":"gpt-4o","model_type":"transcription","capabilities":{"streaming":true},"pricing":{"input":2.5,"output":10},"last_updated":"2026-06-29","description":"GPT-4o Transcribe is a speech-to-text model that uses GPT-4o to transcribe audio.","tagline":"Speech-to-text model powered by GPT-4o","context_window":16000,"max_output_tokens":2000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false,"modalities":{"input":["audio","text"],"output":["text"]}},{"id":"openai/gpt-4o","name":"gpt-4o","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-4o","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"fine_tuning":false,"structured_output":true,"json_mode":true,"reasoning":false,"batch":true},"pricing":{"input":2.5,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-4o (“o” for “omni”) is our versatile, high-intelligence flagship model.","tagline":"Fast, intelligent, flexible GPT model","max_output_tokens":16384,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-chat","name":"gpt-5-chat","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5-codex","name":"gpt-5-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5 optimized for agentic coding in Codex","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-mini","name":"gpt-5-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.25,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 mini is a faster, more cost-efficient version of GPT-5. It's great for","tagline":"Near-frontier intelligence for cost sensitive, low latency, high volume workloads","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-nano","name":"gpt-5-nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.05,"output":0.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 Nano is our fastest, cheapest version of GPT-5. It's great for","tagline":"Fastest, most cost-efficient version of GPT-5","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-05","license":"proprietary","open_weight":false},{"id":"openai/gpt-5-pro","name":"gpt-5-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":120},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex-max","name":"gpt-5.1-codex-max","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT‑5.1-Codex-Max is purpose-built for agentic coding.","tagline":"A version of GPT-5.1-codex optimized for long running tasks.","max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex-mini","name":"gpt-5.1-codex-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":0.25,"output":2},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1-Codex mini is a smaller, more cost-effective, less-capable version of GPT-5.1-Codex.","tagline":"Smaller, more cost-effective, less-capable version of GPT-5.1-Codex","max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-codex","name":"gpt-5.1-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.1-Codex is a version of GPT-5 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.","tagline":"A version of GPT-5.1 optimized for agentic coding in Codex.","max_output_tokens":128000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.1-instant","name":"gpt-5.1-instant","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.1-thinking","name":"gpt-5.1-thinking","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.1","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.2-chat","name":"gpt-5.2-chat","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.2-codex","name":"gpt-5.2-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2-Codex is an upgraded version of GPT-5.2 optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.2-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.2-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"Our most intelligent coding model optimized for long-horizon, agentic coding tasks.","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2-pro","name":"gpt-5.2-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false},"pricing":{"input":21,"output":168},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.2","name":"gpt-5.2","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.2","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.2 is our previous frontier model for complex professional work.","tagline":"Previous frontier model for professional work with configurable reasoning effort","max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.3-chat","name":"gpt-5.3-chat","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-5.3-codex","name":"gpt-5.3-codex","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.3","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false},"pricing":{"input":1.75,"output":14},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.3-Codex is optimized for agentic coding tasks in [Codex](https://developers.openai.com/codex) or similar environments.\\nGPT-5.3-Codex supports `low`, `medium`, `high`, and `xhigh` reasoning effort settings.\\nIf you want to learn more about prompting GPT-5.3-Codex, refer to our [dedicated guide](/cookbook/examples/gpt-5/codex_prompting_guide).\\n","tagline":"The most capable agentic coding model to date.","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-mini","name":"gpt-5.4-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.75,"output":4.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 mini brings the strengths of GPT-5.4 to a faster, more efficient","tagline":"Our strongest mini model yet for coding, computer use, and subagents","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-nano","name":"gpt-5.4-nano","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":0.2,"output":1.25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 nano is designed for tasks where speed and cost matter most like","tagline":"Our cheapest GPT-5.4-class model for simple high-volume tasks","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4-pro","name":"gpt-5.4-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":1100000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false},"pricing":{"input":30,"output":180},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.4","name":"gpt-5.4","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5.4","context_window":1100000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2.5,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5.4 is our frontier model for complex professional work.","tagline":"A more affordable model for coding and professional work.","max_output_tokens":128000,"knowledge_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5-pro","name":"gpt-5.5-pro","created_by":"openai","source":"official","family":"gpt-5.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":30,"output":180},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5.5","name":"gpt-5.5","created_by":"openai","source":"official","family":"gpt-5.5","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":5,"output":30},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-25","description":"GPT-5.5 is our newest frontier model for the most complex professional work.","tagline":"A new class of intelligence for coding and professional work.","max_output_tokens":128000,"knowledge_cutoff":"2025-12","license":"proprietary","open_weight":false},{"id":"openai/gpt-5","name":"gpt-5","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-5","context_window":400000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.25,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"GPT-5 is our previous model for coding, reasoning, and agentic tasks across domains.","tagline":"Previous intelligent reasoning model for coding and agentic tasks with configurable reasoning effort","max_output_tokens":128000,"max_input_tokens":272000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false},{"id":"openai/gpt-image-1-mini","name":"gpt-image-1-mini","created_by":"openai","source":"official","last_updated":"2026-03-21","family":"gpt-image-1","capabilities":{"streaming":true},"pricing":{"input":2,"output":8},"model_type":"image","description":"A cost-efficient version of GPT Image 1. It is a natively multimodal language model that accepts both text and image inputs, and produces image outputs.","tagline":"A cost-efficient version of GPT Image 1","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"openai/gpt-image-1.5","name":"gpt-image-1.5","created_by":"openai","source":"official","last_updated":"2026-03-21","family":"gpt-image-1.5","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":5,"output":32},"model_type":"image","description":"GPT Image 1.5 is our previous image generation model, with better instruction following and adherence to prompts. Learn more in our [GPT Image 1.5 usage guide](/api/docs/guides/image-generation).","tagline":"Our previous image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image","text"]}},{"id":"openai/gpt-image-1","name":"gpt-image-1","created_by":"openai","source":"official","last_updated":"2026-03-21","family":"gpt-image-1","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":5,"output":40},"model_type":"image","description":"GPT Image 1 is a natively multimodal language model that accepts both text and image inputs, and produces image outputs.","tagline":"Our previous image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"openai/gpt-image-2","name":"gpt-image-2","created_by":"openai","source":"official","family":"gpt-image-2","model_type":"image","capabilities":{"streaming":true,"vision":false,"tool_call":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false,"batch":true},"pricing":{"input":5,"output":30},"last_updated":"2026-04-22","description":"GPT Image 2 is our state-of-the-art image generation model for fast, high-quality image generation and editing. It supports flexible image sizes and high-fidelity image inputs. Learn more in our [image generation guide](/api/docs/guides/image-generation), or see the [pricing page](/api/docs/pricing#image-generation) and [image generation calculator](/api/docs/guides/image-generation#calculating-costs) for cost estimates.","tagline":"State-of-the-art image generation model","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"openai/gpt-oss-120b","name":"gpt-oss-120b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"pricing":{"input":0.35,"output":0.75},"model_type":"chat","parameters":120,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Most powerful open-weight model, fits into an H100 GPU","max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-20b","name":"gpt-oss-20b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"fine_tuning":false,"json_mode":true,"vision":false,"batch":true},"pricing":{"input":0.07,"output":0.3},"model_type":"chat","parameters":20,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"\\","tagline":"Medium-sized open-weight model for low latency","max_output_tokens":131072,"knowledge_cutoff":"2024-06","license":"apache-2.0","open_weight":true},{"id":"openai/gpt-oss-safeguard-20b","name":"gpt-oss-safeguard-20b","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"gpt-oss","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.07,"output":0.3},"model_type":"chat","parameters":20,"tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/gpt-realtime-1.5","name":"gpt-realtime-1.5","created_by":"openai","source":"official","family":"gpt-realtime","model_type":"audio","capabilities":{"streaming":true,"tool_call":true,"vision":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":4,"output":16},"last_updated":"2026-06-29","description":"GPT-Realtime-1.5 is our flagship audio model for voice agents and customer support.","tagline":"The best voice model for audio in, audio out","context_window":32000,"max_output_tokens":4096,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false,"modalities":{"input":["text","audio","image"],"output":["text","audio"]}},{"id":"openai/gpt-realtime-2","name":"gpt-realtime-2","created_by":"openai","source":"official","family":"gpt-realtime","model_type":"audio","capabilities":{"streaming":true,"vision":false,"tool_call":true,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":true},"pricing":{"input":4,"output":24},"last_updated":"2026-06-29","description":"GPT-Realtime-2 is our most capable realtime voice model. It supports speech-to-speech interactions with configurable reasoning effort, stronger instruction following, and more reliable tool use for complex voice-agent workflows.","tagline":"Reasoning model for realtime voice interactions","context_window":128000,"max_output_tokens":32000,"knowledge_cutoff":"2024-09","license":"proprietary","open_weight":false,"modalities":{"input":["text","audio","image"],"output":["text","audio"]}},{"id":"openai/gpt-realtime-mini","name":"gpt-realtime-mini","created_by":"openai","source":"official","family":"gpt-realtime","model_type":"audio","capabilities":{"streaming":true,"tool_call":true,"vision":false,"structured_output":false,"json_mode":false,"fine_tuning":false,"reasoning":false},"pricing":{"input":0.6,"output":2.4},"last_updated":"2026-06-29","description":"GPT-Realtime mini is capable of responding to audio and text inputs in realtime over WebRTC, WebSocket, or SIP connections.","tagline":"A cost-efficient version of GPT-Realtime","context_window":32000,"max_output_tokens":4096,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false,"modalities":{"input":["text","image","audio"],"output":["text","audio"]}},{"id":"openai/o1","name":"o1","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o1","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":15,"output":60},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The o1 series of models are trained with reinforcement learning to perform complex reasoning. o1 models think before they answer, producing a long internal chain of thought before responding to the user.","tagline":"Previous full o-series reasoning model","max_output_tokens":100000,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-deep-research","name":"o3-deep-research","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":false,"json_mode":false,"fine_tuning":false,"batch":true},"pricing":{"input":10,"output":40},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o3-deep-research is our most advanced model for deep research,","tagline":"Our most powerful deep research model","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3-mini","name":"o3-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":1.1,"output":4.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o3-mini is our newest small reasoning model, providing high intelligence at the same cost and latency targets of o1-mini. o3-mini supports key developer features, like Structured Outputs, function calling, and Batch API.","tagline":"A small model alternative to o3","max_output_tokens":100000,"knowledge_cutoff":"2023-10","license":"proprietary","open_weight":false},{"id":"openai/o3-pro","name":"o3-pro","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":20,"output":80},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"The o-series of models are trained with reinforcement learning to think","tagline":"Version of o3 with more compute for better responses","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o3","name":"o3","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o3","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":false,"batch":true},"pricing":{"input":2,"output":8},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o3 is a well-rounded and powerful model across domains. It sets a new","tagline":"Reasoning model for complex tasks, succeeded by GPT-5","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/o4-mini","name":"o4-mini","created_by":"openai","source":"official","last_updated":"2026-03-23","family":"o4","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"fine_tuning":true,"json_mode":true,"batch":true},"pricing":{"input":1.1,"output":4.4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"o4-mini is our latest small o-series model. It's optimized for fast,","tagline":"Fast, cost-efficient reasoning model, succeeded by GPT-5 mini","max_output_tokens":100000,"knowledge_cutoff":"2024-06","license":"proprietary","open_weight":false},{"id":"openai/text-embedding-3-large","name":"text-embedding-3-large","created_by":"openai","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.13,"output":0},"family":"text-embedding","description":"text-embedding-3-large is our most capable embedding model for both english and non-english tasks.","tagline":"Most capable embedding model","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/text-embedding-3-small","name":"text-embedding-3-small","created_by":"openai","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"family":"text-embedding","description":"text-embedding-3-small is our improved, more performant version of our ada embedding model.","tagline":"Small embedding model","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/text-embedding-ada-002","name":"text-embedding-ada-002","created_by":"openai","source":"official","last_updated":"2026-03-21","model_type":"embed","capabilities":{"streaming":true},"pricing":{"input":0.1,"output":0},"family":"text-embedding","description":"text-embedding-ada-002 is our improved, more performant version of our ada embedding model.","tagline":"Older embedding model","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["text"]}},{"id":"openai/tts-1-hd","name":"tts-1-hd","created_by":"openai","source":"official","family":"tts","model_type":"tts","capabilities":{"streaming":true},"pricing":{"input":30,"output":0},"last_updated":"2026-06-29","description":"TTS is a model that converts text to natural sounding spoken text. The tts-1-hd model is optimized for high quality text-to-speech use cases. Use it with the Speech endpoint in the Audio API.","tagline":"Text-to-speech model optimized for quality","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["audio"]}},{"id":"openai/tts-1","name":"tts-1","created_by":"openai","source":"official","family":"tts","model_type":"tts","capabilities":{"streaming":true},"pricing":{"input":15,"output":0},"last_updated":"2026-06-29","description":"TTS is a model that converts text to natural sounding spoken text. The tts-1 model is optimized for realtime text-to-speech use cases. Use it with the Speech endpoint in the Audio API.","tagline":"Text-to-speech model optimized for speed","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["audio"]}},{"id":"openai/whisper-1","name":"whisper-1","created_by":"openai","source":"official","family":"whisper","model_type":"transcription","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-06-29","description":"Whisper is a general-purpose speech recognition model, trained on a large dataset of diverse audio. You can also use it as a multitask model to perform multilingual speech recognition as well as speech translation and language identification.","tagline":"General-purpose speech recognition model","license":"proprietary","open_weight":false,"modalities":{"input":["audio"],"output":["text"]}},{"id":"perplexity/sonar-pro","name":"sonar-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","context_window":200000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0,"output":0},"family":"sonar","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"perplexity/sonar-reasoning-pro","name":"sonar-reasoning-pro","created_by":"perplexity","source":"official","last_updated":"2026-03-23","context_window":127000,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":0,"output":0},"model_type":"chat","family":"sonar","modalities":{"input":["text"],"output":["text"]}},{"id":"perplexity/sonar-reasoning","name":"sonar-reasoning","created_by":"perplexity","source":"official","last_updated":"2026-03-21","context_window":127000,"capabilities":{"streaming":true,"reasoning":true},"pricing":{"input":1,"output":5},"model_type":"reasoning","family":"sonar"},{"id":"perplexity/sonar","name":"sonar","created_by":"perplexity","source":"official","last_updated":"2026-03-23","context_window":127000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":0,"output":0},"family":"sonar","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"prime-intellect/intellect-3","name":"intellect-3","created_by":"prime-intellect","source":"official","last_updated":"2026-03-23","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":1.1},"tools":["function_calling"],"model_type":"chat","modalities":{"input":["text"],"output":["text"]}},{"id":"prodia/flux-fast-schnell","name":"flux-fast-schnell","created_by":"prodia","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image"},{"id":"quiverai/arrow-1.1","name":"arrow-1.1","created_by":"quiverai","source":"official","context_window":131100,"model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-06-29"},{"id":"recraft/recraft-v2","name":"recraft-v2","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"First-generation Recraft raster model.","tagline":"Legacy Recraft V2 raster.","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["image"]}},{"id":"recraft/recraft-v3","name":"recraft-v3","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Previously codenamed Red Panda; raster model with strong text rendering and brand-style preservation. Used for image-to-image, inpainting, and background editing endpoints.","tagline":"Recraft V3 raster (formerly Red Panda).","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"recraft/recraft-v4-pro","name":"recraft-v4-pro","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"High-fidelity 2048x2048 raster tier of Recraft V4 with improved anatomy and realism for print-grade output.","tagline":"Print-grade raster Recraft V4.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"recraft/recraft-v4.1-pro","name":"recraft-v4.1-pro","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4.1-utility-pro","name":"recraft-v4.1-utility-pro","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4.1-utility","name":"recraft-v4.1-utility","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4.1","name":"recraft-v4.1","created_by":"recraft","source":"official","model_type":"image","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-05-17"},{"id":"recraft/recraft-v4","name":"recraft-v4","created_by":"recraft","source":"official","last_updated":"2026-03-21","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"model_type":"image","description":"Latest Recraft raster model with strong design sensibility, readable text, and consistent composition. Default model in the API.","tagline":"Design-forward raster image model.","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"sakana/fugu-ultra","name":"fugu-ultra","created_by":"sakana","source":"official","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":5,"output":30},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"stepfun/step-3.5-flash","name":"step-3.5-flash","created_by":"stepfun","source":"official","context_window":262100,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"fine_tuning":true},"pricing":{"input":0.09,"output":0.3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02","description":"Flagship reasoning model with 256k context, MoE architecture (199B total, ~11B active). Optimized for agentic tasks, tool calling, planning, math, and coding.","tagline":"Frontier reasoning and agentic capabilities at low cost.","license":"apache-2.0","open_weight":true},{"id":"stepfun/step-3.7-flash","name":"step-3.7-flash","created_by":"stepfun","source":"official","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":1.15},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"voyage/rerank-2.5-lite","name":"rerank-2.5-lite","created_by":"voyage","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/rerank-2.5","name":"rerank-2.5","created_by":"voyage","source":"official","family":"rerank","context_window":32000,"model_type":"rerank","capabilities":{"streaming":true},"pricing":{"input":0.05,"output":0},"last_updated":"2026-04-16","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-3-large","name":"voyage-3-large","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.18,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-3.5-lite","name":"voyage-3.5-lite","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-3.5","name":"voyage-3.5","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-4-large","name":"voyage-4-large","created_by":"voyage","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-4-lite","name":"voyage-4-lite","created_by":"voyage","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.02,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-4","name":"voyage-4","created_by":"voyage","source":"official","last_updated":"2026-03-23","context_window":32000,"capabilities":{"streaming":true},"pricing":{"input":0.06,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-code-2","name":"voyage-code-2","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-code-3","name":"voyage-code-3","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.18,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-finance-2","name":"voyage-finance-2","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"voyage/voyage-law-2","name":"voyage-law-2","created_by":"voyage","source":"official","last_updated":"2026-03-23","capabilities":{"streaming":true},"pricing":{"input":0.12,"output":0},"model_type":"embed","modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-2-vision","name":"grok-2-vision","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-2","context_window":32800,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":2,"output":10},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-3-fast","name":"grok-3-fast","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":5,"output":25},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-3-mini-fast","name":"grok-3-mini-fast","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true},"pricing":{"input":0.6,"output":4},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-3-mini","name":"grok-3-mini","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"json_mode":true,"batch":true},"pricing":{"input":0.3,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"A lightweight model that thinks before responding.","tagline":"A lightweight model that thinks before responding.","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-3","name":"grok-3","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-3","context_window":131100,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Grok 3 is the latest model from xAI.","tagline":"Grok 3 is the latest model from xAI.","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4-fast-non-reasoning","name":"grok-4-fast-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-04-17","family":"grok-4","context_window":2000000,"capabilities":{"streaming":true,"tool_call":true,"vision":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4-fast-reasoning","name":"grok-4-fast-reasoning","created_by":"xai","source":"official","last_updated":"2026-04-17","family":"grok-4","context_window":2000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4.1-fast-non-reasoning","name":"grok-4.1-fast-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-17","family":"grok-4.1","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"vision":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.1-fast-reasoning","name":"grok-4.1-fast-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-17","family":"grok-4.1","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":0.2,"output":0.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-multi-agent-beta","name":"grok-4.20-multi-agent-beta","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":1.25,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-multi-agent","name":"grok-4.20-multi-agent","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"vision":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"xai/grok-4.20-non-reasoning-beta","name":"grok-4.20-non-reasoning-beta","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":1.25,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-non-reasoning","name":"grok-4.20-non-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"xai/grok-4.20-reasoning-beta","name":"grok-4.20-reasoning-beta","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":2.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-4.20-reasoning","name":"grok-4.20-reasoning","created_by":"xai","source":"official","last_updated":"2026-05-06","family":"grok-4.20","context_window":2000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"]},{"id":"xai/grok-4.3","name":"grok-4.3","created_by":"xai","source":"official","family":"grok-4.3","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":1.25,"output":2.5},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-01","knowledge_cutoff":"2024-11","license":"proprietary","open_weight":false},{"id":"xai/grok-4","name":"grok-4","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-4","context_window":256000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":15},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"xai/grok-build-0.1","name":"grok-build-0.1","created_by":"xai","source":"official","context_window":256000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"batch":true,"structured_output":true,"json_mode":true},"pricing":{"input":1,"output":2},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-23","license":"proprietary","open_weight":false},{"id":"xai/grok-code-fast-1","name":"grok-code-fast-1","created_by":"xai","source":"official","last_updated":"2026-03-23","family":"grok-code","context_window":256000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"pricing":{"input":0.2,"output":1.5},"model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]},"description":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","tagline":"Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding.","max_output_tokens":10000,"license":"proprietary","open_weight":false},{"id":"xai/grok-imagine-image-pro","name":"grok-imagine-image-pro","created_by":"xai","source":"official","last_updated":"2026-03-21","family":"grok-imagine","capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"model_type":"image","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"xai/grok-imagine-image","name":"grok-imagine-image","created_by":"xai","source":"official","last_updated":"2026-03-21","family":"grok-imagine","capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"model_type":"image","license":"proprietary","open_weight":false,"modalities":{"input":["text","image"],"output":["image"]}},{"id":"xai/grok-imagine-video-1.5-preview","name":"grok-imagine-video-1.5-preview","created_by":"xai","source":"official","family":"grok-imagine","model_type":"image","capabilities":{"streaming":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-06-03","license":"proprietary","open_weight":false,"modalities":{"input":["image"],"output":["video"]}},{"id":"xai/grok-imagine-video-1.5","name":"grok-imagine-video-1.5","created_by":"xai","source":"official","family":"grok-imagine","model_type":"image","capabilities":{"streaming":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"last_updated":"2026-06-29","license":"proprietary","open_weight":false,"modalities":{"input":["image"],"output":["video"]}},{"id":"xai/grok-imagine-video","name":"grok-imagine-video","created_by":"xai","source":"official","last_updated":"2026-03-21","family":"grok-imagine","capabilities":{"streaming":true,"tool_call":true,"batch":true,"vision":true},"pricing":{"input":0,"output":0},"model_type":"image","license":"proprietary","open_weight":false,"modalities":{"input":["text","image","video"],"output":["video"]}},{"id":"xai/grok-stt","name":"grok-stt","created_by":"xai","source":"official","model_type":"transcription","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"last_updated":"2026-06-29"},{"id":"xai/grok-tts","name":"grok-tts","created_by":"xai","source":"official","model_type":"tts","capabilities":{"streaming":true},"pricing":{"input":15,"output":0},"last_updated":"2026-06-29"},{"id":"xai/grok-voice-think-fast-1.0","name":"grok-voice-think-fast-1.0","created_by":"xai","source":"official","model_type":"chat","capabilities":{"streaming":true},"pricing":{"input":0,"output":0},"modalities":{"input":["text"],"output":["text"]},"last_updated":"2026-06-29"},{"id":"xiaomi/mimo-v2-flash","name":"mimo-v2-flash","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","context_window":262100,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"fine_tuning":true,"structured_output":true},"pricing":{"input":0.1,"output":0.3},"tools":["function_calling"],"family":"mimo","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"309B MoE reasoning model with 15B active parameters. Fast inference at 150 tokens/s.","tagline":"309B MoE reasoning model with 15B active parameters.","max_output_tokens":64000,"knowledge_cutoff":"2024-12","license":"mit","parameters":309,"active_parameters":15,"architecture":"moe","open_weight":true},{"id":"xiaomi/mimo-v2-pro","name":"mimo-v2-pro","created_by":"xiaomi","source":"official","last_updated":"2026-03-23","context_window":1000000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true,"json_mode":true,"structured_output":true},"pricing":{"input":1,"output":3},"tools":["function_calling"],"family":"mimo","model_type":"chat","modalities":{"input":["text"],"output":["text"]},"description":"Flagship foundation model with 1M-token context and top-tier agentic capabilities.","tagline":"Flagship foundation model with 1M-token context and top-tier agentic capabilities.","max_output_tokens":128000,"license":"proprietary","parameters":1000,"active_parameters":42,"architecture":"moe","open_weight":false},{"id":"xiaomi/mimo-v2.5-pro","name":"mimo-v2.5-pro","created_by":"xiaomi","source":"official","family":"mimo","context_window":1100000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.43,"output":0.87},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","max_output_tokens":128000},{"id":"xiaomi/mimo-v2.5","name":"mimo-v2.5","created_by":"xiaomi","source":"official","family":"mimo","context_window":1100000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true},"pricing":{"input":0.14,"output":0.28},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-05-29","max_output_tokens":128000},{"id":"zai/glm-4.5-air","name":"glm-4.5-air","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.2,"output":1.1},"family":"glm-4.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.5","name":"glm-4.5","created_by":"zhipu","source":"official","last_updated":"2026-06-29","context_window":128000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":2.2},"family":"glm-4.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.5v","name":"glm-4.5v","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":66000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":1.8},"family":"glm-4.5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.6","name":"glm-4.6","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.6,"output":2.2},"family":"glm-4.6","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.6v-flash","name":"glm-4.6v-flash","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0,"output":0},"family":"glm-4.6","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.6v","name":"glm-4.6v","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.3,"output":0.9},"family":"glm-4.6","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.7-flash","name":"glm-4.7-flash","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":200000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.07,"output":0.4},"family":"glm-4.7","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.7-flashx","name":"glm-4.7-flashx","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":200000,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.06,"output":0.4},"family":"glm-4.7","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-4.7","name":"glm-4.7","created_by":"zhipu","source":"official","last_updated":"2026-04-07","context_window":204800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":2.25,"output":2.75},"family":"glm-4.7","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-5-turbo","name":"glm-5-turbo","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.2,"output":4},"family":"glm-5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-5.1","name":"glm-5.1","created_by":"zhipu","source":"official","family":"glm-5.1","context_window":204800,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.3,"output":4.3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-02"},{"id":"zai/glm-5.2-fast","name":"glm-5.2-fast","created_by":"zhipu","source":"official","family":"glm-5.2","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":3,"output":10.25},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"zai/glm-5.2","name":"glm-5.2","created_by":"zhipu","source":"official","family":"glm-5.2","context_window":1000000,"model_type":"chat","capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.95,"output":3},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-06-29"},{"id":"zai/glm-5","name":"glm-5","created_by":"zhipu","source":"official","last_updated":"2026-03-23","context_window":202800,"capabilities":{"streaming":true,"tool_call":true,"reasoning":true},"pricing":{"input":0.8,"output":2.56},"family":"glm-5","model_type":"chat","tools":["function_calling"],"modalities":{"input":["text"],"output":["text"]}},{"id":"zai/glm-5v-turbo","name":"glm-5v-turbo","created_by":"zhipu","source":"official","family":"glm-5","context_window":200000,"model_type":"chat","capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true},"pricing":{"input":1.2,"output":4},"modalities":{"input":["text"],"output":["text"]},"tools":["function_calling"],"last_updated":"2026-04-03"}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
diff --git a/packages/npm/src/providers/vertex.ts b/packages/npm/src/providers/vertex.ts
index ca76cc2d5..cd109298d 100644
--- a/packages/npm/src/providers/vertex.ts
+++ b/packages/npm/src/providers/vertex.ts
@@ -3,6 +3,6 @@
import type { ProviderWithModels, Model } from '../types';
-export const provider: ProviderWithModels = {"id":"vertex","name":"Google Vertex AI","region":"US","headquarters":"Mountain View, CA","url":"https://cloud.google.com/vertex-ai","api_url":"https://us-central1-aiplatform.googleapis.com/v1","docs_url":"https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models","pricing_url":"https://cloud.google.com/vertex-ai/generative-ai/pricing","description":"Google Cloud Vertex AI provides enterprise access to Gemini, Claude, Llama, and Mistral models.","type":"cloud","playground_url":"https://console.cloud.google.com/vertex-ai/generative","status_url":"https://status.cloud.google.com","sdk":{"python":"google-cloud-aiplatform","javascript":"@google-cloud/vertexai"},"openai_compatible":true,"free_tier":false,"models_url":"https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models","twitter_url":"https://x.com/GoogleDeepMind","blog_url":"https://cloud.google.com/blog/products/ai-machine-learning","terms_url":"https://cloud.google.com/terms/service-terms","support_url":"https://cloud.google.com/vertex-ai/docs/support/getting-support","icon":"","models":[{"id":"antigravity-preview-05-2026","name":"antigravity-preview-05-2026","created_by":"google","source":"official","status":"active","context_window":1048576,"max_output_tokens":65536,"license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-23"},{"id":"claude-haiku-4-5@vertex","name":"Claude Haiku 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"max_output_tokens":64000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.1,"output":5.5,"cached_input":0.11,"batch_input":0.55,"batch_output":2.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.1,0.11,5.5]},{"label":"Batch","values":[0.55,null,2.75]}]}]},"model_type":"chat","tools":["function_calling"],"description":"The fastest model with near-frontier intelligence","tagline":"The fastest model with near-frontier intelligence","knowledge_cutoff":"2025-02","training_data_cutoff":"2025-07","license":"proprietary","open_weight":false},{"id":"claude-opus-4-6@vertex","name":"Claude Opus 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":1000000,"max_output_tokens":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5.5,"output":27.5,"cached_input":0.55,"batch_input":2.75,"batch_output":13.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[5.5,0.55,27.5]},{"label":"Batch","values":[2.75,null,13.75]}]}]},"model_type":"chat","tools":["function_calling"],"description":"The most intelligent broadly available model for agents and coding","tagline":"The most intelligent model for building agents and coding","knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4-6@vertex","name":"Claude Sonnet 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"max_output_tokens":64000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":3.3,"output":16.5,"cached_input":0.33,"batch_input":1.65,"batch_output":8.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[3.3,0.33,16.5]},{"label":"Batch","values":[1.65,null,8.25]}]}]},"model_type":"chat","tools":["function_calling"],"description":"The best combination of speed and intelligence","tagline":"The best combination of speed and intelligence","knowledge_cutoff":"2025-08","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"deep-research-max-preview-04-2026","name":"deep-research-max-preview-04-2026","created_by":"google","source":"official","family":"deep-research","status":"active","context_window":1048576,"max_output_tokens":65536,"model_type":"reasoning","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-22"},{"id":"deep-research-preview-04-2026","name":"deep-research-preview-04-2026","created_by":"google","source":"official","family":"deep-research","status":"active","context_window":1048576,"max_output_tokens":65536,"model_type":"reasoning","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-22"},{"id":"deep-research-pro-preview-12-2025","name":"deep-research-pro-preview-12-2025","created_by":"google","source":"official","last_updated":"2026-04-22","status":"active","context_window":1048576,"max_output_tokens":65536,"model_type":"reasoning","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"deep-research","license":"proprietary","open_weight":false},{"id":"embedding-001","name":"embedding-001","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"embed","license":"proprietary","open_weight":false},{"id":"embedding-gecko-001","name":"embedding-gecko-001","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"embed","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-001","name":"gemini-2.0-flash-001","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.5-flash"},{"id":"gemini-2.0-flash-lite-001","name":"gemini-2.0-flash-lite-001","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.1-flash-lite"},{"id":"gemini-2.0-flash-lite-preview-02-05","name":"gemini-2.0-flash-lite-preview-02-05","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-2.5-flash-lite","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-lite-preview","name":"gemini-2.0-flash-lite-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-2.5-flash-lite","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-lite","name":"gemini-2.0-flash-lite","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","knowledge_cutoff":"2024-08","context_window":1048576,"max_output_tokens":8192,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":0.075,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.075,null,0.3]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.1-flash-lite"},{"id":"gemini-2.0-flash-live-001","name":"gemini-2.0-flash-live-001","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-3.1-flash-live-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-preview-image-generation","name":"gemini-2.0-flash-preview-image-generation","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-11-14","successor":"gemini-2.5-flash-image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash","name":"gemini-2.0-flash","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","knowledge_cutoff":"2024-08","context_window":1048576,"max_output_tokens":8192,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.5-flash"},{"id":"gemini-2.5-computer-use-preview-10-2025","name":"gemini-2.5-computer-use-preview-10-2025","created_by":"google","source":"official","last_updated":"2026-04-22","family":"gemini-2.5","status":"active","context_window":128000,"max_output_tokens":64000,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-image-preview","name":"gemini-2.5-flash-image-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2026-01-15","successor":"gemini-2.5-flash-image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-image","name":"gemini-2.5-flash-image","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-06","context_window":65536,"max_output_tokens":32768,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency. Gemini 2.5 Flash Image, also known as Nano Banana, is best for high-volume generation, conversational image editing, and low-latency creative workflows that require native multimodal understanding.","tagline":"Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency."},{"id":"gemini-2.5-flash-lite-preview-09-2025","name":"gemini-2.5-flash-lite-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-05-10","family":"gemini-2.5","status":"deprecated","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-03-31","successor":"gemini-3.1-flash-lite"},{"id":"gemini-2.5-flash-lite","name":"gemini-2.5-flash-lite","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.1,"output":0.4,"batch_input":0.05,"batch_output":0.2,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.1,null,0.4]},{"label":"Priority","values":[0.18,null,0.72]},{"label":"Flex","values":[0.05,null,0.2]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 2.5 Flash-Lite is best for high-volume classification, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks."},{"id":"gemini-2.5-flash-native-audio-preview-12-2025","name":"gemini-2.5-flash-native-audio-preview-12-2025","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":8192,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Live API enables low-latency, real-time voice and video interactions with Gemini 2.5 Flash. It processes continuous streams of audio, video, or text to deliver immediate, human-like spoken responses, creating a natural conversational experience for your users.","tagline":"The Live API enables low-latency, real-time voice and video interactions with Gemini 2.5 Flash."},{"id":"gemini-2.5-flash-preview-05-20","name":"gemini-2.5-flash-preview-05-20","created_by":"google","source":"official","last_updated":"2026-05-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-11-18","successor":"gemini-3.5-flash","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-preview-09-2025","name":"gemini-2.5-flash-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-preview-09-25","name":"gemini-2.5-flash-preview-09-25","created_by":"google","source":"official","last_updated":"2026-05-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2026-02-17","successor":"gemini-3.5-flash","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-preview-tts","name":"gemini-2.5-flash-preview-tts","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","status":"active","context_window":8192,"max_output_tokens":16384,"model_type":"tts","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"license":"proprietary","open_weight":false,"description":"Our fastest engine for high-fidelity speech synthesis, offering low-latency and cost-efficient audio generation. Gemini 2.5 Flash TTS is best for real-time assistants, high-volume narration, and conversational use cases that require fine-grained control over voice style and pacing.","tagline":"Our fastest engine for high-fidelity speech synthesis, offering low-latency and cost-efficient audio generation."},{"id":"gemini-2.5-flash","name":"gemini-2.5-flash","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"Our best model in terms of price-performance, offering well-rounded capabilities. 2.5 Flash is best for large scale processing, low-latency, high volume tasks that require thinking, and agentic use cases.","tagline":"Our best model in terms of price-performance, offering well-rounded capabilities."},{"id":"gemini-2.5-pro-preview-03-25","name":"gemini-2.5-pro-preview-03-25","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-12-02","successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-pro-preview-05-06","name":"gemini-2.5-pro-preview-05-06","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-12-02","successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-pro-preview-06-05","name":"gemini-2.5-pro-preview-06-05","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-12-02","successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-pro-preview-tts","name":"gemini-2.5-pro-preview-tts","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","status":"active","context_window":8192,"max_output_tokens":16384,"model_type":"tts","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"license":"proprietary","open_weight":false,"description":"Our premium engine for studio-quality speech synthesis, offering high-fidelity and sophisticated audio generation. Gemini 2.5 Pro TTS is best for long-form content, professional narrations, and complex creative workflows that require the highest level of vocal clarity and natural prosody.","tagline":"Our premium engine for studio-quality speech synthesis, offering high-fidelity and sophisticated audio generation."},{"id":"gemini-2.5-pro","name":"gemini-2.5-pro","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","tagline":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context."},{"id":"gemini-3-flash-preview","name":"gemini-3-flash-preview","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-3","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.5,"output":3,"batch_input":0.25,"batch_output":1.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.5,null,3]},{"label":"Priority","values":[0.9,null,5.4]},{"label":"Flex","values":[0.25,null,1.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of state-of-the-art reasoning.","tagline":"The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of stat"},{"id":"gemini-3-pro-image-preview","name":"gemini-3-pro-image-preview","created_by":"google","source":"official","last_updated":"2026-05-29","family":"gemini-3","status":"active","knowledge_cutoff":"2025-01","context_window":65536,"max_output_tokens":32768,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-3-pro-image","name":"gemini-3-pro-image","created_by":"google","source":"official","family":"gemini-3","status":"active","knowledge_cutoff":"2025-01","context_window":65536,"max_output_tokens":32768,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-29"},{"id":"gemini-3-pro-preview","name":"gemini-3-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3","status":"deprecated","deprecation_date":"2026-03-09","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[3.6,null,21.6]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-3.1-flash-image-preview","name":"gemini-3.1-flash-image-preview","created_by":"google","source":"official","last_updated":"2026-05-29","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":32768,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":3,"batch_input":0.25,"batch_output":1.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.5,null,3]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[0.25,null,1.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-3.1-flash-image","name":"gemini-3.1-flash-image","created_by":"google","source":"official","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":32768,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":0.5,"output":3,"batch_input":0.25,"batch_output":1.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.5,null,3]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[0.25,null,1.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-29"},{"id":"gemini-3.1-flash-lite-preview","name":"gemini-3.1-flash-lite-preview","created_by":"google","source":"official","last_updated":"2026-05-29","family":"gemini-3.1","status":"deprecated","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":1.5,"batch_input":0.125,"batch_output":0.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.25,null,1.5]},{"label":"Priority","values":[0.45,null,2.7]},{"label":"Flex","values":[0.125,null,0.75]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 3.1 Flash-Lite is best for high-volume agentic tasks, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","deprecation_date":"2026-05-25","successor":"gemini-3.1-flash-lite"},{"id":"gemini-3.1-flash-lite","name":"gemini-3.1-flash-lite","created_by":"google","source":"official","family":"gemini-3.1","description":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary constraints.","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","tagline":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary con","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":1.5,"batch_input":0.125,"batch_output":0.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.25,null,1.5]},{"label":"Priority","values":[0.45,null,2.7]},{"label":"Flex","values":[0.125,null,0.75]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-10"},{"id":"gemini-3.1-flash-live-preview","name":"gemini-3.1-flash-live-preview","created_by":"google","source":"official","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-03-27"},{"id":"gemini-3.1-flash-tts-preview","name":"gemini-3.1-flash-tts-preview","created_by":"google","source":"official","family":"gemini-3.1","description":"The Gemini 3.1 Flash TTS Preview model provides powerful, low-latency speech generation with natural outputs, steerable prompts, and new expressive audio tags for precise narration control.","status":"active","knowledge_cutoff":"2025-01","context_window":8192,"max_output_tokens":16384,"model_type":"tts","license":"proprietary","tagline":"The Gemini 3.1 Flash TTS Preview model provides powerful, low-latency speech generation with natural outputs, steerable prompts, and new expressive audio tags for precise narration control.","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-16"},{"id":"gemini-3.1-pro-preview","name":"gemini-3.1-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[3.6,null,21.6]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-3.5-flash","name":"gemini-3.5-flash","created_by":"google","source":"official","family":"gemini-3.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":1.5,"output":9,"batch_input":0.75,"batch_output":4.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.5,null,9]},{"label":"Priority","values":[2.7,null,16.2]},{"label":"Flex","values":[0.75,null,4.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-23"},{"id":"gemini-3.5-live-translate-preview","name":"gemini-3.5-live-translate-preview","created_by":"google","source":"official","family":"gemini-3.5","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":65536,"model_type":"translation","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-10"},{"id":"gemini-embedding-001","name":"gemini-embedding-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","context_window":2048,"model_type":"embed","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-embedding-2-preview","name":"gemini-embedding-2-preview","created_by":"google","source":"official","last_updated":"2026-03-23","status":"active","context_window":8192,"model_type":"embed","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-embedding-2","name":"gemini-embedding-2","created_by":"google","source":"official","family":"gemini-embedding","status":"active","context_window":8192,"model_type":"embed","license":"proprietary","open_weight":false,"capabilities":{"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-24"},{"id":"gemini-embedding-exp-03-07","name":"gemini-embedding-exp-03-07","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-embedding-exp","name":"gemini-embedding-exp","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-live-2.5-flash-preview","name":"gemini-live-2.5-flash-preview","created_by":"google","source":"official","last_updated":"2026-03-27","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-3.1-flash-live-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"family":"gemini-2.5","model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-robotics-er-1.5-preview","name":"gemini-robotics-er-1.5-preview","created_by":"google","source":"official","last_updated":"2026-05-01","status":"deprecated","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"gemini-1.5","model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-04-30","successor":"gemini-robotics-er-1.6-preview"},{"id":"gemini-robotics-er-1.6-preview","name":"gemini-robotics-er-1.6-preview","created_by":"google","source":"official","family":"gemini-1.6","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-01"},{"id":"imagen-3.0-generate-002","name":"imagen-3.0-generate-002","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2025-11-10","model_type":"image","successor":"imagen-4.0-generate-001","modalities":{"input":["text"],"output":["image"]},"family":"imagen-3.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-fast-generate-001","name":"imagen-4.0-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","model_type":"image","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-generate-001","name":"imagen-4.0-generate-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","model_type":"image","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-generate-preview-06-06","name":"imagen-4.0-generate-preview-06-06","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2026-02-17","model_type":"image","successor":"imagen-4.0-generate-001","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-ultra-generate-001","name":"imagen-4.0-ultra-generate-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","model_type":"image","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-ultra-generate-preview-06-06","name":"imagen-4.0-ultra-generate-preview-06-06","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2026-02-17","model_type":"image","successor":"imagen-4.0-ultra-generate-001","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen","name":"imagen","created_by":"google","source":"official","last_updated":"2026-06-18","status":"active","context_window":480,"model_type":"image","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["image"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"imagen","license":"proprietary","open_weight":false,"description":"The Imagen 4 standard, ultra, and fast endpoints are deprecated and will be shut down on **August 17, 2026**; migrate to Gemini 3.1 Flash Image to avoid service interruptions. A high-performance engine for advanced visual synthesis, offering sophisticated creative control and photorealistic output. Use Imagen 4 for professional branding, intricate scene composition, and high-fidelity design tasks that require precise text rendering and complex lighting.","tagline":"A high-performance engine for advanced visual synthesis, offering sophisticated creative control and photorealistic output."},{"id":"jamba-1.5-large@vertex","name":"Jamba 1.5 Large","created_by":"ai21","source":"official","last_updated":"2026-03-22","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat"},{"id":"jamba-1.5-mini@vertex","name":"Jamba 1.5 Mini","created_by":"ai21","source":"official","last_updated":"2026-03-22","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat"},{"id":"lyria-3-clip-preview","name":"lyria-3-clip-preview","created_by":"google","source":"official","status":"active","context_window":131072,"model_type":"tts","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-03-26"},{"id":"lyria-3-pro-preview","name":"lyria-3-pro-preview","created_by":"google","source":"official","status":"active","context_window":131072,"model_type":"tts","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-03-26"},{"id":"lyria-realtime-exp","name":"lyria-realtime-exp","created_by":"google","source":"official","last_updated":"2026-03-24","status":"active","model_type":"tts","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"license":"proprietary","open_weight":false},{"id":"mistral-medium-3@vertex","name":"Mistral Medium 3","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-medium","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.4,null,2]}]}]},"model_type":"chat","tools":["function_calling"]},{"id":"text-embedding-004","name":"text-embedding-004","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2026-01-14","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"text-embedding","license":"proprietary","open_weight":false},{"id":"veo-2.0-generate-001","name":"veo-2.0-generate-001","created_by":"google","source":"official","last_updated":"2026-06-18","status":"active","model_type":"video","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["video"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"veo-2.0","license":"proprietary","open_weight":false,"description":"A models available through the Gemini Enterprise Agent Platform to avoid service interruptions. Our robust and reliable stable video generation model, providing consistent performance for production-ready workflows. Veo 2.0 is best for standard 1080p content creation, social media assets, and high-volume video tasks that prioritize reliable character consistency and predictable sub-second generation speeds.","tagline":"Our robust and reliable stable video generation model, providing consistent performance for production-ready workflows."},{"id":"veo-3.0-fast-generate-001","name":"veo-3.0-fast-generate-001","created_by":"google","source":"official","family":"veo-3.0","status":"active","model_type":"video","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]},"last_updated":"2026-06-18"},{"id":"veo-3.0-fast-generate-preview","name":"veo-3.0-fast-generate-preview","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2025-11-12","model_type":"video","successor":"veo-3.1-fast-generate-preview","modalities":{"input":["text"],"output":["video"]},"family":"veo-3.0","license":"proprietary","open_weight":false},{"id":"veo-3.0-generate-001","name":"veo-3.0-generate-001","created_by":"google","source":"official","family":"veo-3.0","status":"active","model_type":"video","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]},"last_updated":"2026-06-18"},{"id":"veo-3.0-generate-preview","name":"veo-3.0-generate-preview","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2025-11-12","model_type":"video","successor":"veo-3.1-generate-preview","modalities":{"input":["text"],"output":["video"]},"family":"veo-3.0","license":"proprietary","open_weight":false},{"id":"veo-3.1-fast-generate-preview","name":"veo-3.1-fast-generate-preview","created_by":"google","source":"official","family":"veo-3.1","status":"deprecated","deprecation_date":"2025-10-15","model_type":"video","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]},"last_updated":"2026-03-26"},{"id":"veo-3.1-generate-preview","name":"veo-3.1-generate-preview","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","context_window":1024,"model_type":"video","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["video"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"veo-3.1","license":"proprietary","open_weight":false,"description":"A state-of-the-art cinematic engine designed for high-end creative storytelling and experimental video production. Veo 3.1 is best for professional-grade 4K output, natively synchronized audio generation, and complex camera movements that require the highest level of temporal consistency and artistic control.","tagline":"A state-of-the-art cinematic engine designed for high-end creative storytelling and experimental video production."},{"id":"veo-3.1-lite-generate-preview","name":"veo-3.1-lite-generate-preview","created_by":"google","source":"official","family":"veo-3.1","status":"active","context_window":1024,"model_type":"video","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-01"}]};
+export const provider: ProviderWithModels = {"id":"vertex","name":"Google Vertex AI","region":"US","headquarters":"Mountain View, CA","url":"https://cloud.google.com/vertex-ai","api_url":"https://us-central1-aiplatform.googleapis.com/v1","docs_url":"https://docs.cloud.google.com/vertex-ai/generative-ai/docs/models","pricing_url":"https://cloud.google.com/vertex-ai/generative-ai/pricing","description":"Google Cloud Vertex AI provides enterprise access to Gemini, Claude, Llama, and Mistral models.","type":"cloud","playground_url":"https://console.cloud.google.com/vertex-ai/generative","status_url":"https://status.cloud.google.com","sdk":{"python":"google-cloud-aiplatform","javascript":"@google-cloud/vertexai"},"openai_compatible":true,"free_tier":false,"models_url":"https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models","twitter_url":"https://x.com/GoogleDeepMind","blog_url":"https://cloud.google.com/blog/products/ai-machine-learning","terms_url":"https://cloud.google.com/terms/service-terms","support_url":"https://cloud.google.com/vertex-ai/docs/support/getting-support","icon":"","models":[{"id":"antigravity-preview-05-2026","name":"antigravity-preview-05-2026","created_by":"google","source":"official","status":"active","context_window":1048576,"max_output_tokens":65536,"license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-05-23"},{"id":"claude-haiku-4-5@vertex","name":"Claude Haiku 4.5","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-haiku","context_window":200000,"max_output_tokens":64000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.1,"output":5.5,"cached_input":0.11,"batch_input":0.55,"batch_output":2.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.1,0.11,5.5]},{"label":"Batch","values":[0.55,null,2.75]}]}]},"model_type":"chat","tools":["function_calling"],"description":"The fastest model with near-frontier intelligence","tagline":"The fastest model with near-frontier intelligence","knowledge_cutoff":"2025-02","training_data_cutoff":"2025-07","license":"proprietary","open_weight":false},{"id":"claude-opus-4-6@vertex","name":"Claude Opus 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-opus","context_window":1000000,"max_output_tokens":128000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":5.5,"output":27.5,"cached_input":0.55,"batch_input":2.75,"batch_output":13.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[5.5,0.55,27.5]},{"label":"Batch","values":[2.75,null,13.75]}]}]},"model_type":"chat","tools":["function_calling"],"description":"The most intelligent broadly available model for agents and coding","tagline":"The most intelligent model for building agents and coding","knowledge_cutoff":"2025-05","training_data_cutoff":"2025-08","license":"proprietary","open_weight":false},{"id":"claude-sonnet-4-6@vertex","name":"Claude Sonnet 4.6","created_by":"anthropic","source":"official","last_updated":"2026-03-23","family":"claude-sonnet","context_window":1000000,"max_output_tokens":64000,"capabilities":{"streaming":true,"vision":true,"tool_call":true,"reasoning":true,"structured_output":true,"json_mode":true,"batch":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":3.3,"output":16.5,"cached_input":0.33,"batch_input":1.65,"batch_output":8.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[3.3,0.33,16.5]},{"label":"Batch","values":[1.65,null,8.25]}]}]},"model_type":"chat","tools":["function_calling"],"description":"The best combination of speed and intelligence","tagline":"The best combination of speed and intelligence","knowledge_cutoff":"2025-08","training_data_cutoff":"2026-01","license":"proprietary","open_weight":false},{"id":"deep-research-max-preview-04-2026","name":"deep-research-max-preview-04-2026","created_by":"google","source":"official","family":"deep-research","status":"active","context_window":1048576,"max_output_tokens":65536,"model_type":"reasoning","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-22"},{"id":"deep-research-preview-04-2026","name":"deep-research-preview-04-2026","created_by":"google","source":"official","family":"deep-research","status":"active","context_window":1048576,"max_output_tokens":65536,"model_type":"reasoning","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-22"},{"id":"deep-research-pro-preview-12-2025","name":"deep-research-pro-preview-12-2025","created_by":"google","source":"official","last_updated":"2026-04-22","status":"active","context_window":1048576,"max_output_tokens":65536,"model_type":"reasoning","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"deep-research","license":"proprietary","open_weight":false},{"id":"embedding-001","name":"embedding-001","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"embed","license":"proprietary","open_weight":false},{"id":"embedding-gecko-001","name":"embedding-gecko-001","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"embed","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-001","name":"gemini-2.0-flash-001","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.5-flash"},{"id":"gemini-2.0-flash-lite-001","name":"gemini-2.0-flash-lite-001","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.1-flash-lite"},{"id":"gemini-2.0-flash-lite-preview-02-05","name":"gemini-2.0-flash-lite-preview-02-05","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-2.5-flash-lite","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-lite-preview","name":"gemini-2.0-flash-lite-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-2.5-flash-lite","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-lite","name":"gemini-2.0-flash-lite","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","knowledge_cutoff":"2024-08","context_window":1048576,"max_output_tokens":8192,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":0.075,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.075,null,0.3]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.1-flash-lite"},{"id":"gemini-2.0-flash-live-001","name":"gemini-2.0-flash-live-001","created_by":"google","source":"official","last_updated":"2026-03-27","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-3.1-flash-live-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash-preview-image-generation","name":"gemini-2.0-flash-preview-image-generation","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.0","status":"deprecated","deprecation_date":"2025-11-14","successor":"gemini-2.5-flash-image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.0-flash","name":"gemini-2.0-flash","created_by":"google","source":"official","last_updated":"2026-06-02","family":"gemini-2.0","status":"deprecated","knowledge_cutoff":"2024-08","context_window":1048576,"max_output_tokens":8192,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":0.15,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.15,null,0.6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-01","successor":"gemini-3.5-flash"},{"id":"gemini-2.5-computer-use-preview-10-2025","name":"gemini-2.5-computer-use-preview-10-2025","created_by":"google","source":"official","last_updated":"2026-04-22","family":"gemini-2.5","status":"active","context_window":128000,"max_output_tokens":64000,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-image-preview","name":"gemini-2.5-flash-image-preview","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2026-01-15","successor":"gemini-2.5-flash-image","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-image","name":"gemini-2.5-flash-image","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-06","context_window":65536,"max_output_tokens":32768,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Image (Nano Banana) Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency. Gemini 2.5 Flash Image, also known as Nano Banana, is best for high-volume generation, conversational image editing, and low-latency creative workflows that require native multimodal understanding.","tagline":"Our best engine for high-velocity visual creation, offering state-of-the-art speed and efficiency."},{"id":"gemini-2.5-flash-lite-preview-09-2025","name":"gemini-2.5-flash-lite-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-05-10","family":"gemini-2.5","status":"deprecated","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-03-31","successor":"gemini-3.1-flash-lite"},{"id":"gemini-2.5-flash-lite","name":"gemini-2.5-flash-lite","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.1,"output":0.4,"batch_input":0.05,"batch_output":0.2,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.1,null,0.4]},{"label":"Priority","values":[0.18,null,0.72]},{"label":"Flex","values":[0.05,null,0.2]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash-Lite Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 2.5 Flash-Lite is best for high-volume classification, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks."},{"id":"gemini-2.5-flash-native-audio-preview-12-2025","name":"gemini-2.5-flash-native-audio-preview-12-2025","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":8192,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Live Preview The Live API enables low-latency, real-time voice and video interactions with Gemini 2.5 Flash. It processes continuous streams of audio, video, or text to deliver immediate, human-like spoken responses, creating a natural conversational experience for your users.","tagline":"The Live API enables low-latency, real-time voice and video interactions with Gemini 2.5 Flash."},{"id":"gemini-2.5-flash-preview-05-20","name":"gemini-2.5-flash-preview-05-20","created_by":"google","source":"official","last_updated":"2026-05-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-11-18","successor":"gemini-3.5-flash","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-preview-09-2025","name":"gemini-2.5-flash-preview-09-2025","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-preview-09-25","name":"gemini-2.5-flash-preview-09-25","created_by":"google","source":"official","last_updated":"2026-05-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2026-02-17","successor":"gemini-3.5-flash","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-flash-preview-tts","name":"gemini-2.5-flash-preview-tts","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","context_window":8192,"max_output_tokens":16384,"model_type":"tts","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Text-to-Speech Our fastest engine for high-fidelity speech synthesis, offering low-latency and cost-efficient audio generation. Gemini 2.5 Flash TTS is best for real-time assistants, high-volume narration, and conversational use cases that require fine-grained control over voice style and pacing.","tagline":"Our fastest engine for high-fidelity speech synthesis, offering low-latency and cost-efficient audio generation."},{"id":"gemini-2.5-flash","name":"gemini-2.5-flash","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.3,"output":2.5,"batch_input":0.15,"batch_output":1.25,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.3,null,2.5]},{"label":"Priority","values":[0.54,null,4.5]},{"label":"Flex","values":[0.15,null,1.25]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Flash Our best model in terms of price-performance, offering well-rounded capabilities. 2.5 Flash is best for large scale processing, low-latency, high volume tasks that require thinking, and agentic use cases.","tagline":"Our best model in terms of price-performance, offering well-rounded capabilities."},{"id":"gemini-2.5-pro-preview-03-25","name":"gemini-2.5-pro-preview-03-25","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-12-02","successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-pro-preview-05-06","name":"gemini-2.5-pro-preview-05-06","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-12-02","successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-pro-preview-06-05","name":"gemini-2.5-pro-preview-06-05","created_by":"google","source":"official","last_updated":"2026-03-23","family":"gemini-2.5","status":"deprecated","deprecation_date":"2025-12-02","successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-2.5-pro-preview-tts","name":"gemini-2.5-pro-preview-tts","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","context_window":8192,"max_output_tokens":16384,"model_type":"tts","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Pro Text-to-Speech Our premium engine for studio-quality speech synthesis, offering high-fidelity and sophisticated audio generation. Gemini 2.5 Pro TTS is best for long-form content, professional narrations, and complex creative workflows that require the highest level of vocal clarity and natural prosody.","tagline":"Our premium engine for studio-quality speech synthesis, offering high-fidelity and sophisticated audio generation."},{"id":"gemini-2.5-pro","name":"gemini-2.5-pro","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-2.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":1.25,"output":10,"batch_input":0.625,"batch_output":5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.25,null,10]},{"label":"Priority","values":[2.25,null,18]},{"label":"Flex","values":[0.625,null,5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 2.5 Pro Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context.","tagline":"Our state-of-the-art thinking model, capable of reasoning over complex problems in code, math, and STEM, as well as analyzing large datasets, codebases, and documents using long context."},{"id":"gemini-3-flash-preview","name":"gemini-3-flash-preview","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-3","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.5,"output":3,"batch_input":0.25,"batch_output":1.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.5,null,3]},{"label":"Priority","values":[0.9,null,5.4]},{"label":"Flex","values":[0.25,null,1.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3 Flash Preview The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of state-of-the-art reasoning.","tagline":"The best model in the world for multimodal understanding, and our most powerful agentic and vibe-coding model yet, delivering richer visuals and deeper interactivity, all built on a foundation of stat"},{"id":"gemini-3-pro-image-preview","name":"gemini-3-pro-image-preview","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-3","status":"deprecated","knowledge_cutoff":"2025-01","context_window":65536,"max_output_tokens":32768,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-25","successor":"gemini-3-pro-image"},{"id":"gemini-3-pro-image","name":"gemini-3-pro-image","created_by":"google","source":"official","family":"gemini-3","status":"active","knowledge_cutoff":"2025-01","context_window":65536,"max_output_tokens":32768,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3 Pro Image Nano Banana Pro is a sophisticated reasoning-driven engine for professional-grade image editing and generation, offering studio-quality precision and advanced creative control. Nano Banana Pro is best for complex graphic design, high-fidelity product mockups, and factual data visualizations that require accurate text rendering and real-world grounding via Google Search.","tagline":"The Interactions API is now generally available."},{"id":"gemini-3-pro-preview","name":"gemini-3-pro-preview","created_by":"google","source":"official","last_updated":"2026-03-24","family":"gemini-3","status":"deprecated","deprecation_date":"2026-03-09","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"successor":"gemini-3.1-pro-preview","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[3.6,null,21.6]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-3.1-flash-image-preview","name":"gemini-3.1-flash-image-preview","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-3.1","status":"deprecated","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":32768,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["text"]},"pricing":{"input":0.5,"output":3,"batch_input":0.25,"batch_output":1.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.5,null,3]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[0.25,null,1.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-06-25","successor":"gemini-3.1-flash-image"},{"id":"gemini-3.1-flash-image","name":"gemini-3.1-flash-image","created_by":"google","source":"official","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":32768,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"pricing":{"input":0.5,"output":3,"batch_input":0.25,"batch_output":1.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.5,null,3]},{"label":"Priority","values":[null,null,null]},{"label":"Flex","values":[0.25,null,1.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash Image Nano Banana 2 provides high-quality image generation and conversational editing at a mainstream price point and low latency. It serves as the high-efficiency counterpart to Gemini 3 Pro Image , optimized for speed and high-volume developer use cases. Key updates: New output resolution options: New support for 0.5K, 2K and 4K, default 1K New Image Search Grounding: Integration of both text and image search results to inform generation with real-time web data Supported with Thinking on or off New 1:4, 4:1, 1:8 and 8:1 aspect ratios Improved aspect ratio adherence Improved image quality and consistency Improved i18n text rendering","tagline":"The Interactions API is now generally available."},{"id":"gemini-3.1-flash-lite-preview","name":"gemini-3.1-flash-lite-preview","created_by":"google","source":"official","last_updated":"2026-05-29","family":"gemini-3.1","status":"deprecated","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":1.5,"batch_input":0.125,"batch_output":0.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.25,null,1.5]},{"label":"Priority","values":[0.45,null,2.7]},{"label":"Flex","values":[0.125,null,0.75]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks. Gemini 3.1 Flash-Lite is best for high-volume agentic tasks, simple data extraction, and extremely low-latency applications where budget and speed are the primary constraints.","tagline":"Our most cost-efficient multimodal model, offering the fastest performance for high-frequency, lightweight tasks.","deprecation_date":"2026-05-25","successor":"gemini-3.1-flash-lite"},{"id":"gemini-3.1-flash-lite","name":"gemini-3.1-flash-lite","created_by":"google","source":"official","family":"gemini-3.1","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash-Lite Gemini 3.1 Flash-Lite is a low-latency, cost-effective multimodal model optimized for high-frequency, lightweight tasks. The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary constraints.","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","tagline":"The model supports text, image, video, audio, and PDF inputs, and is designed for high-volume agentic workflows, simple data extraction, and applications where latency and API cost are the primary con","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":0.25,"output":1.5,"batch_input":0.125,"batch_output":0.75,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.25,null,1.5]},{"label":"Priority","values":[0.45,null,2.7]},{"label":"Flex","values":[0.125,null,0.75]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29"},{"id":"gemini-3.1-flash-live-preview","name":"gemini-3.1-flash-live-preview","created_by":"google","source":"official","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash Live Preview Gemini 3.1 Flash Live Preview is our low-latency, audio-to-audio model optimized for real-time dialogue and voice-first AI applications with acoustic nuance detection, numeric precision, and multimodal awareness.","tagline":"The Interactions API is now generally available."},{"id":"gemini-3.1-flash-tts-preview","name":"gemini-3.1-flash-tts-preview","created_by":"google","source":"official","family":"gemini-3.1","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Flash TTS (Text-to-Speech) Preview The Gemini 3.1 Flash TTS Preview model provides powerful, low-latency speech generation with natural outputs, steerable prompts, and new expressive audio tags for precise narration control.","status":"active","knowledge_cutoff":"2025-01","context_window":8192,"max_output_tokens":16384,"model_type":"tts","license":"proprietary","tagline":"The Gemini 3.1 Flash TTS Preview model provides powerful, low-latency speech generation with natural outputs, steerable prompts, and new expressive audio tags for precise narration control.","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29"},{"id":"gemini-3.1-pro-preview","name":"gemini-3.1-pro-preview","created_by":"google","source":"official","last_updated":"2026-06-29","family":"gemini-3.1","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":2,"output":12,"batch_input":1,"batch_output":6,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[2,null,12]},{"label":"Priority","values":[3.6,null,21.6]},{"label":"Flex","values":[1,null,6]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"model_type":"chat","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.1 Pro Preview Built to refine the performance and reliability of the Gemini 3 Pro series, Gemini 3.1 Pro Preview provides better thinking, improved token efficiency, and a more grounded, factually consistent experience. It's optimized for software engineering behavior and usability, as well as agentic workflows requiring precise tool usage and reliable multi-step execution across real-world domains.","tagline":"The Interactions API is now generally available."},{"id":"gemini-3.5-flash","name":"gemini-3.5-flash","created_by":"google","source":"official","family":"gemini-3.5","status":"active","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"pricing":{"input":1.5,"output":9,"batch_input":0.75,"batch_output":4.5,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[1.5,null,9]},{"label":"Priority","values":[2.7,null,16.2]},{"label":"Flex","values":[0.75,null,4.5]}]}]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.5 Flash Gemini 3.5 Flash provides sustained frontier-level intelligence optimized for real-world tasks at a higher speed and lower cost. Designed for the agentic era, it excels at sub-agent deployment, multi-step workflows, and long-horizon tasks at scale. This model is particularly effective for rapid agentic loops involving complex coding cycles and iterations.","tagline":"The Interactions API is now generally available."},{"id":"gemini-3.5-live-translate-preview","name":"gemini-3.5-live-translate-preview","created_by":"google","source":"official","family":"gemini-3.5","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":65536,"model_type":"translation","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini 3.5 Live Translate Gemini 3.5 Live Translate is our low-latency, audio-to-audio model optimized for real-time translation of spoken conversations. It enables seamless, bidirectional translation with high accuracy and natural voice output.","tagline":"The Interactions API is now generally available."},{"id":"gemini-embedding-001","name":"gemini-embedding-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","context_window":2048,"model_type":"embed","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-embedding-2-preview","name":"gemini-embedding-2-preview","created_by":"google","source":"official","last_updated":"2026-03-23","status":"active","context_window":8192,"model_type":"embed","capabilities":{"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-embedding-2","name":"gemini-embedding-2","created_by":"google","source":"official","family":"gemini-embedding","status":"active","context_window":8192,"model_type":"embed","license":"proprietary","open_weight":false,"capabilities":{"vision":true},"modalities":{"input":["text"],"output":["text"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-04-24"},{"id":"gemini-embedding-exp-03-07","name":"gemini-embedding-exp-03-07","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-embedding-exp","name":"gemini-embedding-exp","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2025-10-30","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"gemini-embedding","license":"proprietary","open_weight":false},{"id":"gemini-live-2.5-flash-preview","name":"gemini-live-2.5-flash-preview","created_by":"google","source":"official","last_updated":"2026-03-27","status":"deprecated","deprecation_date":"2025-12-09","successor":"gemini-3.1-flash-live-preview","capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["text"]},"family":"gemini-2.5","model_type":"chat","license":"proprietary","open_weight":false},{"id":"gemini-robotics-er-1.5-preview","name":"gemini-robotics-er-1.5-preview","created_by":"google","source":"official","last_updated":"2026-06-29","status":"deprecated","knowledge_cutoff":"2025-01","context_window":1048576,"max_output_tokens":65536,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"gemini-1.5","model_type":"chat","license":"proprietary","open_weight":false,"deprecation_date":"2026-04-30","successor":"gemini-robotics-er-1.6-preview","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini Robotics-ER 1.5 Note: This version is planned for deprecation soon . See the models page for the latest model version. Gemini Robotics-ER 1.5 is a vision-language model (VLM) that brings Gemini's agentic capabilities to robotics. It's designed for advanced reasoning in the physical world, allowing robots to interpret complex visual data, perform spatial reasoning, and plan actions from natural language commands.","tagline":"The Interactions API is now generally available."},{"id":"gemini-robotics-er-1.6-preview","name":"gemini-robotics-er-1.6-preview","created_by":"google","source":"official","family":"gemini-1.6","status":"active","knowledge_cutoff":"2025-01","context_window":131072,"max_output_tokens":65536,"model_type":"chat","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image","video","audio"],"output":["text","image","audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Gemini Robotics-ER 1.6 Gemini Robotics-ER 1.6 is a vision-language model (VLM) that brings Gemini's agentic capabilities to robotics. It's designed for advanced reasoning in the physical world, allowing robots to interpret complex visual data, perform spatial reasoning, and plan actions from natural language commands.","tagline":"The Interactions API is now generally available."},{"id":"imagen-3.0-generate-002","name":"imagen-3.0-generate-002","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2025-11-10","model_type":"image","successor":"imagen-4.0-generate-001","modalities":{"input":["text"],"output":["image"]},"family":"imagen-3.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-fast-generate-001","name":"imagen-4.0-fast-generate-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","model_type":"image","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-generate-001","name":"imagen-4.0-generate-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","model_type":"image","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-generate-preview-06-06","name":"imagen-4.0-generate-preview-06-06","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2026-02-17","model_type":"image","successor":"imagen-4.0-generate-001","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-ultra-generate-001","name":"imagen-4.0-ultra-generate-001","created_by":"google","source":"official","last_updated":"2026-03-27","status":"active","model_type":"image","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen-4.0-ultra-generate-preview-06-06","name":"imagen-4.0-ultra-generate-preview-06-06","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2026-02-17","model_type":"image","successor":"imagen-4.0-ultra-generate-001","modalities":{"input":["text"],"output":["image"]},"family":"imagen-4.0","license":"proprietary","open_weight":false},{"id":"imagen","name":"imagen","created_by":"google","source":"official","last_updated":"2026-06-29","status":"active","context_window":480,"model_type":"image","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["image"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"imagen","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Imagen 4 The Imagen 4 standard, ultra, and fast endpoints are deprecated and will be shut down on **August 17, 2026**; migrate to Gemini 3.1 Flash Image to avoid service interruptions. A high-performance engine for advanced visual synthesis, offering sophisticated creative control and photorealistic output. Use Imagen 4 for professional branding, intricate scene composition, and high-fidelity design tasks that require precise text rendering and complex lighting.","tagline":"A high-performance engine for advanced visual synthesis, offering sophisticated creative control and photorealistic output."},{"id":"jamba-1.5-large@vertex","name":"Jamba 1.5 Large","created_by":"ai21","source":"official","last_updated":"2026-03-22","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat"},{"id":"jamba-1.5-mini@vertex","name":"Jamba 1.5 Mini","created_by":"ai21","source":"official","last_updated":"2026-03-22","capabilities":{"streaming":true},"modalities":{"input":["text"],"output":["text"]},"model_type":"chat"},{"id":"lyria-3-clip-preview","name":"lyria-3-clip-preview","created_by":"google","source":"official","status":"active","context_window":131072,"model_type":"tts","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Lyria 3 Clip Preview Lyria 3 Clip Preview is Google's model optimized for generating short musical clips, loops, and previews. It generates 30-second high-quality, 48kHz stereo audio from text prompts or image inputs.","tagline":"The Interactions API is now generally available."},{"id":"lyria-3-pro-preview","name":"lyria-3-pro-preview","created_by":"google","source":"official","status":"active","context_window":131072,"model_type":"tts","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Lyria 3 Pro Preview Lyria 3 Pro Preview is Google's flagship music generation model. It is optimized for generating full-length songs with complex structural coherence, including multiple verses, choruses, and bridges. It generates high-quality, 48kHz stereo audio from text prompts or image inputs.","tagline":"The Interactions API is now generally available."},{"id":"lyria-realtime-exp","name":"lyria-realtime-exp","created_by":"google","source":"official","last_updated":"2026-03-24","status":"active","model_type":"tts","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text"],"output":["audio"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"license":"proprietary","open_weight":false},{"id":"mistral-medium-3@vertex","name":"Mistral Medium 3","created_by":"mistral","source":"official","last_updated":"2026-03-23","family":"mistral-medium","capabilities":{"streaming":true,"tool_call":true},"modalities":{"input":["text"],"output":["text"]},"pricing":{"input":0.4,"output":2,"tiers":[{"label":"Text tokens","unit":"Per 1M tokens","columns":["Input","Cached input","Output"],"rows":[{"label":"Standard","values":[0.4,null,2]}]}]},"model_type":"chat","tools":["function_calling"]},{"id":"text-embedding-004","name":"text-embedding-004","created_by":"google","source":"official","last_updated":"2026-06-02","status":"deprecated","deprecation_date":"2026-01-14","model_type":"embed","successor":"gemini-embedding-2","modalities":{"input":["text"],"output":["text"]},"family":"text-embedding","license":"proprietary","open_weight":false},{"id":"veo-2.0-generate-001","name":"veo-2.0-generate-001","created_by":"google","source":"official","last_updated":"2026-06-29","status":"active","model_type":"video","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["video"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"veo-2.0","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Veo 2.0 This model is deprecated and will be shut down on **June 30, 2026**; migrate to Veo 3.1 Preview or the GA models available through the Gemini Enterprise Agent Platform to avoid service interruptions. Our robust and reliable stable video generation model, providing consistent performance for production-ready workflows. Veo 2.0 is best for standard 1080p content creation, social media assets, and high-volume video tasks that prioritize reliable character consistency and predictable sub-second generation speeds.","tagline":"Our robust and reliable stable video generation model, providing consistent performance for production-ready workflows."},{"id":"veo-3.0-fast-generate-001","name":"veo-3.0-fast-generate-001","created_by":"google","source":"official","family":"veo-3.0","status":"active","model_type":"video","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]},"last_updated":"2026-06-18"},{"id":"veo-3.0-fast-generate-preview","name":"veo-3.0-fast-generate-preview","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2025-11-12","model_type":"video","successor":"veo-3.1-fast-generate-preview","modalities":{"input":["text"],"output":["video"]},"family":"veo-3.0","license":"proprietary","open_weight":false},{"id":"veo-3.0-generate-001","name":"veo-3.0-generate-001","created_by":"google","source":"official","family":"veo-3.0","status":"active","model_type":"video","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]},"last_updated":"2026-06-18"},{"id":"veo-3.0-generate-preview","name":"veo-3.0-generate-preview","created_by":"google","source":"official","last_updated":"2026-03-23","status":"deprecated","deprecation_date":"2025-11-12","model_type":"video","successor":"veo-3.1-generate-preview","modalities":{"input":["text"],"output":["video"]},"family":"veo-3.0","license":"proprietary","open_weight":false},{"id":"veo-3.1-fast-generate-preview","name":"veo-3.1-fast-generate-preview","created_by":"google","source":"official","family":"veo-3.1","status":"deprecated","deprecation_date":"2025-10-15","model_type":"video","license":"proprietary","open_weight":false,"modalities":{"input":["text"],"output":["video"]},"last_updated":"2026-03-26"},{"id":"veo-3.1-generate-preview","name":"veo-3.1-generate-preview","created_by":"google","source":"official","last_updated":"2026-06-29","status":"active","context_window":1024,"model_type":"video","capabilities":{"streaming":true,"tool_call":true,"structured_output":true,"reasoning":true,"batch":true,"vision":true,"json_mode":true},"modalities":{"input":["text","image"],"output":["video"]},"tools":["function_calling"],"endpoints":["generateContent","streamGenerateContent"],"family":"veo-3.1","license":"proprietary","open_weight":false,"description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Veo 3.1 A state-of-the-art cinematic engine designed for high-end creative storytelling and experimental video production. Veo 3.1 is best for professional-grade 4K output, natively synchronized audio generation, and complex camera movements that require the highest level of temporal consistency and artistic control.","tagline":"A state-of-the-art cinematic engine designed for high-end creative storytelling and experimental video production."},{"id":"veo-3.1-lite-generate-preview","name":"veo-3.1-lite-generate-preview","created_by":"google","source":"official","family":"veo-3.1","status":"active","context_window":1024,"model_type":"video","license":"proprietary","open_weight":false,"capabilities":{"streaming":true,"vision":true},"modalities":{"input":["text","image"],"output":["video"]},"endpoints":["generateContent","streamGenerateContent"],"last_updated":"2026-06-29","description":"The Interactions API is now generally available. We recommend using this API for access to all the latest features and models. Home Gemini API Docs Send feedback Veo 3.1 Lite Preview Veo 3.1 Lite Preview is a high-efficiency, developer-first video model providing high-fidelity video generation, editing, and cinematic control. It leverages the state-of-the-art Veo 3.1 model to democratize professional-grade video AI by offering a scalable, programmable interface for creators and enterprises.","tagline":"The Interactions API is now generally available."}]};
export const models: Model[] = provider.models.map((m) => ({ ...m, provider: provider.id }));
](https://discord.gg/u54VK8m8tk) [](https://ko-fi.com/unsloth) [](https://github.com/unslothai/unsloth)","release_date":"2024-02-14","context_window":2048,"license":"apache-2.0","parameters":1.1,"tagline":"[](https://discord.gg/u54VK8m8tk) [