LLM 0.32rc2 Release
New features and improvements in LLM 0.32rc2, including a new default model and CLI tool
The LLM 0.32rc2 release has been announced, bringing several new features and improvements. The default model for users who have not set their own default is now GPT-5.6 Luna, which is a more recent and better model than the previous default, GPT-4o mini. However, it is slightly more expensive. Users can switch back to the previous default or choose an even cheaper option, GPT-5 nano.
The release also includes a new command, llm openai endpoint, which allows users to run prompts, chats, and model listings against arbitrary OpenAI-compatible endpoints without configuring a model first. This command is not logged.
Why it matters
This release demonstrates the ongoing development and improvement of LLM, with a focus on providing users with more options and flexibility. The new default model and CLI tool can enhance the user experience and provide more accurate results.
This release demonstrates the ongoing development and improvement of LLM, with a focus on providing users with more options and flexibility.
What you can learn from this
- Model selection: The choice of model can significantly impact the results and cost of using LLM. Understanding the differences between models like GPT-5.6 Luna and GPT-4o mini can help users make informed decisions about which model to use. To apply this, consider experimenting with different models and evaluating their performance on your specific tasks.
- CLI tools: The
llm openai endpointcommand demonstrates the importance of having flexible and user-friendly CLI tools. Learners can apply this by exploring other CLI tools and libraries that can simplify their workflow and improve productivity. - OpenAI compatibility: The ability to run prompts and model listings against arbitrary OpenAI-compatible endpoints highlights the value of interoperability in AI systems. To learn from this, consider exploring other AI frameworks and libraries that provide similar compatibility and flexibility.
- Cost and performance tradeoffs: The different pricing models for GPT-5.6 Luna and GPT-4o mini illustrate the tradeoffs between cost and performance in AI systems. Learners can apply this by considering the cost and performance implications of their own AI projects and making informed decisions about resource allocation.
- Community-driven development: The LLM 0.32rc2 release is the result of community-driven development, with contributors adding new features and improving existing ones. To learn from this, consider participating in open-source AI projects and contributing to the development of new features and tools.
We teach this
Sources
- llm 0.32rc2 — Simon Willison
Our reporting is an original summary; full coverage is at the links above.
Don't just read about it — build it.
Square 1 teaches the skills behind the headlines, with every line of your work graded by AI. Find your starting point in 3 minutes.
Get your free skill report