New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging
I released LLM 0.32 this morning, the most significant new version of LLM since the initial launch of the project. The new version includes support for visible reasoning traces, server-side provider tools, redesigned content-addressable SQLite logs, new models, and new features enabled by the OpenAI Responses API. I also released a new version of the llm-anthropic plugin with substantial updates…
On August 4th, 2026, Simon Willison released version 0.32 of the Long-Form Language Model (LLM), a significant update since the project's initial launch. This new version introduces several features including visible reasoning traces, server-side provider tools, redesigned content-addressable SQLite logs, new models, and improvements enabled by the OpenAI Responses API.
A plugin for llm-anthropic, an enhancement of the llm-anthropic plugin, adds WebSearch, WebFetch, CodeExecution, and AnthropicMCP capabilities. These enhancements enable LLM to execute prompts against various API endpoints, including OpenAI compatible endpoints, without the need for an LLM installation.
The Python API of LLM has been revamped to accept a list of messages as input, simplifying complex scenarios by sending all messages at once. This change allows LLM to output a more versatile sequence of reasoning text, tool calls, and even image attachments. The latest release also supports the semi-standard OpenAI chat completions API, with the introduction of the llm-chat-completions-server plugin.
One of the key challenges addressed in this release is logging. To avoid logging duplicate JSON for every message sequence in a prompt, LLM now utilizes a content-addressable message store, akin to Git's design. The llm logs and llm logs --json commands have been upgraded to convert this new message format back into a user-friendly format.
However, existing plugins may require upgrading to 0.32 to take full advantage of the new streaming events system. A comprehensive guide on implementing plugins with Structured messages and streaming events is now available in the documentation.
Written by urgent.news from Simon Willison's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
