Skip to content

sculptdotfun/modelgrep

Repository files navigation

modelgrep — find & understand every LLM

modelgrep

Find and understand every LLM. Compare 300+ models by benchmark, speed, latency, price, context and capability — in one place.

Website API Next.js TypeScript Stars

modelgrep.com · API docs

modelgrep aggregates OpenRouter's catalog with independent benchmarks, live performance and pricing so you can compare large language models by intelligence, speed, latency, cost, context and capability.


What it does

  • Leaderboard — every model, filterable by capability (reasoning, tools, vision, JSON), price, context and provider, sortable on any metric.
  • Rankings — data-backed "best LLM for X" pages: smartest, coding, design, fastest, cheapest, free, reasoning, vision, agents, long-context, open-source — and each scoped per maker (e.g. fastest Anthropic model).
  • Compare — side-by-side model vs. model.
  • Makers & model pages — per-provider breakdowns with live throughput, latency, uptime and per-endpoint pricing.
  • Free public API — the same data as read-only JSON, no key required.

Data sources

Source Provides
OpenRouter Base catalog, pricing, capabilities, providers, live throughput & latency
Artificial Analysis Intelligence / Coding / Agentic index scores (GPQA, SWE-bench, SciCode, Tau²…)
Design Arena Head-to-head Elo for UI / frontend / design generation

Performance refreshes hourly; benchmarks daily. All enrichment is cached so the heavy fan-out runs once per revalidate window.

API

Free, read-only, no API key, CORS-enabled. Base URL: https://modelgrep.com/api/v1

# Cheapest vision models under $1 / M tokens
curl "https://modelgrep.com/api/v1/models?capabilities=vision&max_price=1&sort=price_input&limit=5"

# Best Anthropic model for coding (ranked + one-sentence answer)
curl "https://modelgrep.com/api/v1/rankings/coding/anthropic"
Endpoint Description
GET /api/v1 Self-describing discovery doc
GET /api/v1/models List, filter, sort & paginate the catalog
GET /api/v1/models/{id} Single model + per-provider breakdown
GET /api/v1/rankings/{collection}[/{maker}] "Best LLM for X" ranking
GET /api/v1/makers Makers with counts & headline models

Full reference: modelgrep.com/api

Tech stack

Local development

Requires Node 22 (see .nvmrc) and pnpm.

pnpm install
pnpm dev        # http://localhost:3000
pnpm build      # production build (fetches live catalog)
pnpm start      # serve the production build

No environment variables or API keys are needed — all upstream sources are keyless.

Project layout

app/        Next.js routes — leaderboard, /best, /compare, /makers, /models, /api
components/  UI (Dashboard, ModelTable, AnswerBox, SiteNav…)
lib/         Data layer — openrouter client, catalog cache, collections, facets, api

License

Unofficial, independent project — not affiliated with OpenRouter, Artificial Analysis or Design Arena.

About

Find & compare 300+ LLMs by benchmark, speed, latency, price & capability — plus a free JSON API. Aggregates OpenRouter, Artificial Analysis & Design Arena.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors