Add vLLM AFD Plugin blog post

[jiangkuaixue123]

Summary

• add an announcement-style post introducing the vLLM AFD Plugin
• explain Attention–FFN disaggregation, the request-driven Attention service, the connector-driven FFN service, and the connector architecture
• cover CUDA and Ascend NPU support, including synchronous NPU AFD and asynchronous prefill AFD
• include the current CAM async prefill performance snapshot and roadmap
• add the architecture and performance figures under the site assets directory

Why

The AFD Plugin repository has detailed implementation and runtime documentation, but the vLLM community does not yet have a concise ecosystem-level introduction to its motivation, architecture, current capabilities, limitations, and future direction.

The source draft is also available in JiusiServe/afd-plugin#114 for project-level review.

Review notes

• The synchronous CAMP2pAFDConnector decode performance subsection is intentionally left as a placeholder until benchmark results are available.
• Repository links use the planned vllm-project/afd-plugin canonical location and should be confirmed before publication.
• The post is dated 2026-07-14.

Validation

• parsed and validated the required YAML front matter
• confirmed the SEO summary is 183 characters
• xmllint --noout assets/figures/2026-07-14-vllm-afd-plugin/vllm-afd-plugin-architecture.svg
• verified both referenced site assets exist
• git diff --cached --check

The local Jekyll build could not be run because the host provides Ruby 2.6 and Bundler 1.17, which cannot resolve the repository's current gem dependencies. The pull request's GitHub Actions build should validate the site with the configured Ruby environment.

[jiangkuaixue123]

cc @hsliuustc0106 @youkaichao @ywang96

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c83b2854ab

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Point plugin links at a public repository

This announcement sends readers to https://github.com/vllm-project/afd-plugin for the plugin, install steps, recipes, docs, and issues, but that repository is not publicly reachable right now; exact URL searches do not surface the repo and a direct browser open returns GitHub's missing-repository page. Once this post is published, users will hit broken install/documentation links unless the repo is made public or these URLs are updated to the actual public location.

Useful? React with 👍 / 👎.

[hsliuustc0106]

Suggested change

	* **Broader vLLM compatibility and upstream alignment:** track newer vLLM and vLLM-Ascend releases, evaluate model runner v2, keep compatibility patches minimal, and contribute generally useful abstractions upstream as they mature.
	* **Broader vLLM compatibility and upstream alignment:** track newer vLLM releases, evaluate model runner v2, keep compatibility patches minimal, and contribute generally useful abstractions upstream as they mature.