New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging
LLM 0.32, released on 4th August 2026, is described as the most significant new version since the project's initial launch. For CLI users, running reasoning models now displays reasoning traces to standard error, with a -R/--hide-reasoning flag to disable this. The release includes out-of-the-box support for the GPT-5.6 model family, and the new default model for llm "prompt" is GPT-5.6 Luna. Server-side tools are now supported, including OpenAI's CodeInterpreter and WebSearch, and the llm-anthropic plugin adds WebSearch, WebFetch, CodeExecution, and AnthropicMCP. The new llm openai endpoint command executes prompts against any OpenAI-compatible endpoint as a one-liner, without logging. In the Python API, a new model.prompt(messages=[]) parameter allows sending a complete message history in a single request, and stream_events() provides a way to handle reasoning text, output strings, tool calls, and image attachments. These features enable a robust implementation of the OpenAI chat completions API, released as the llm-chat-completions-server plugin.
LLM 0.32 adds reasoning traces, server-side tools, and a message-list API, enabling more advanced and flexible LLM workflows.