Simon Willison has developed a new server plugin called `llm-chat-completions-server` that allows users to access their local LLM models through an OpenAI Chat Completions-compatible endpoint. This server, written by GPT-5.6 "Sol", supports the OpenAI API structure, enabling conversational requests where the client manages conversation state. The plugin is designed to de-duplicate conversation messages using content-addressable logs, a feature introduced in LLM 0.32rc1. AI
IMPACT Enables broader compatibility and easier integration of local LLMs with existing AI application ecosystems.
RANK_REASON The cluster describes a new software plugin that integrates existing LLM models with a specific API standard, rather than a core model release or significant industry event.
Read on Mastodon — mastodon.social →
- Claude Code
- GPT 5.6 "Sol"
- Kimi k3
- llm-chat-completions-server
- OpenAI
- Python
- Qwen3.5 4B
- Simon Willison
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
