New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging

LLM 0.32 separates displayed reasoning from the answer passed to another program. In Simon Willison's release explanation, the command-line tool sends reasoning traces to standard error, a separate output channel, while keeping answer text on standard output for piping into other tools.

Its Python interface likewise distinguishes text, reasoning and other events instead of treating every response as a sequence of plain strings. That lets applications handle the different parts deliberately. The command-line option -R/--hide-reasoning turns the reasoning display off, and model plugins need updates to participate fully in the new event system.

“I released LLM 0.32 this morning, the most significant new version of LLM since the initial launch of the project.”

Simon Willison

“Running LLM against reasoning models now displays their reasoning traces to standard error, so you can see what they are "thinking" without that information being included in the standard output that you might pipe to another tool.”

Simon Willison

“When I started work on LLM, the term "agent" had such a vague definition that I refused to use it.”

Simon Willison

“Maybe the next version of LLM will bake the concept of an "agent" into the core library. I'm still trying to figure out what that would look like.”

Simon Willison