llm 0.32rc2
Simon Willison · Simon Willison · 2026-07-30
Simon Willison released llm 0.32rc2, updating his open-source LLM CLI tool to default to GPT-5.6 Luna and adding an `llm openai endpoint` command for running prompts against any OpenAI-compatible API endpoint without prior model configuration.
Appears in
Extraction
Topics: llm-cli-toolsopen-source-toolsopenai-compatible-apideveloper-tooling
Claims
- llm 0.32rc2 switches the default model from GPT-4o mini to GPT-5.6 Luna, which costs $0.20/$1.20 per million tokens versus $0.15/$0.60 for 4o mini.
- The new `llm openai endpoint` command allows sending prompts to arbitrary OpenAI Chat Completions-compatible endpoints without configuring a model plugin first.
- Calls made via `llm openai endpoint` are deliberately not logged by the tool.
- The command can be used as a uvx one-liner without installing llm globally, enabling ad-hoc testing of local model servers.
Key quotes
I got frustrated at the lack of an obvious CLI tool for trying out prompts against arbitrary OpenAI Chat Completions imitation endpoints, so I decided to add that to LLM itself.