New LLM Server Brings Chat Completions Compatibility

The LLM project has released a server that provides compatibility with OpenAI's Chat Completions API. Running locally on port 9001, the tool exposes all installed LLM models through a ChatGPT-compatible endpoint, enabling efficient processing of incremental conversations. The new server leverages the redesigned schema from LLM version 0.32rc1 to deduplicate messages using cryptographic hashes, optimizing conversational log storage. This approach ensures that successive requests extending an existing conversation do not create duplicate data. Compatibility with OpenAI standards removes the need for code modifications in projects already using Chat Completions, streamlining integration with the LLM ecosystem.