Key Points

The post announces a new GitHub release (“research-llm-apis 2026-04-04”) containing research into HTTP APIs used by multiple LLM providers.
It explains that Willison is planning a major change to his LLM Python library and CLI, driven by provider-specific capabilities like server-side tool execution that his existing abstraction can’t model well.
Claude Code was used to review the Python client libraries for Anthropic, OpenAI, Gemini, and Mistral and to generate representative `curl` commands for both streaming and non-streaming JSON responses.
The generated scripts and captured API outputs are included in the new repository to support designing an updated abstraction layer.

Simon Willison’s Weblog

Sponsored by: WorkOS — Production-ready APIs for auth and access control, so you can ship faster.

5th April 2026

Release research-llm-apis 2026-04-04 — Research into the HTTP APIs from various LLM providers.

I'm working on a major change to my LLM Python library and CLI tool. LLM provides an abstraction layer over hundreds of different LLMs from dozens of different vendors thanks to its plugin system, and some of those vendors have grown new features over the past year which LLM's abstraction layer can't handle, such as server-side tool execution.

To help design that new abstraction layer I had Claude Code read through the Python client libraries for Anthropic, OpenAI, Gemini and Mistral and use those to help craft curl commands to access the raw JSON for both streaming and non-streaming modes across a range of different scenarios. Both the scripts and the captured outputs now live in this new repo.

Posted 5th April 2026 at 12:32 am