release
llm 0.31.1
For developers building AI workflows that rely on tool calling, this fix prevents unexpected errors when tools receive empty arguments, ensuring smoother integrations with various LLM providers.
What happened
Simon Willison released version 0.31.1 of the command-line tool `llm`, addressing a bug that caused JSON errors when using OpenAI-compatible chat completion endpoints. The issue occurred when a tool call included empty arguments, which some providers could not handle correctly. Willison discovered the bug while testing the `llm-meta-ai` plugin. This release is a minor patch that ensures robust handling of edge cases in tool calling, a feature increasingly used to give LLMs structured output and function execution capabilities.
Key takeaways
- Version 0.31.1 of the `llm` tool was released to fix a bug with tool calls having empty arguments.
- The bug triggered JSON errors on certain providers' OpenAI Chat Completion endpoints.
- The issue was discovered by Simon Willison during testing of the `llm-meta-ai` plugin.
- The fix improves reliability for developers integrating tool calling into their LLM workflows.
Why it matters
For developers building AI workflows that rely on tool calling, this fix prevents unexpected errors when tools receive empty arguments, ensuring smoother integrations with various LLM providers.
This is an original editorial digest by AI Workflow Center. Full reporting at the source:
Read the original on Simon WillisonMore AI news
All news →
Run Your Own AI Directory