The Information Machine

llm-chat-completions-server 0.1a0

Simon Willison · Simon Willison · 2026-07-30

Simon Willison released llm-chat-completions-server, an LLM plugin that runs a local server exposing all installed LLM models through an OpenAI Chat Completions-compatible API, written entirely by GPT-5.6 Sol.

Open original ↗

Appears in

Extraction

Topics: llm-cli-toolsopenai-compatible-apilocal-model-servingdeveloper-tooling

Claims

  • The llm-chat-completions-server plugin starts a local HTTP server that exposes all LLM-installed models via an OpenAI Chat Completions-compatible endpoint.
  • The plugin was built to validate the content-addressable conversation logging in LLM 0.32, which de-duplicates multi-turn messages using hashes of individual message parts.
  • GPT-5.6 Sol wrote the entire plugin autonomously, demonstrating sufficient knowledge of the OpenAI Chat Completions API to produce a working server implementation.
  • The server enables any OpenAI-compatible client application to route requests to locally managed models without additional configuration.

Key quotes

GPT-5.6 Sol wrote the whole thing - it turns out it knows the OpenAI Chat Completions API shape really well.