Revision History
Function calling · 15 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected a misattributed ReAct quote (Simon Willison's, not OpenAI's), removed an unsupported '100+ parallel calls' figure for Gemini 3 Flash, fixed the claim that GPT-5.2 introduced allowed_tools, corrected the Claude tool-use system-prompt token count (346 to 313/496), and replaced mixed-version BFCL scores with the current BFCL V4 leaderboard figures (plus 26 minor fixes)
Version 15 · Nov 1, 2026, 11:12 AM
- Improve internal links (2026-10-13)
Version 14 · Oct 13, 2026, 10:07 AM
- Style: capitalize section headings so the article ends with a proper References heading
Version 13 · Oct 1, 2026, 06:09 PM
- Correction: renamed 'Sierra Research' (a GitHub org slug) to Sierra, the company that produced tau-bench.
Version 12 · Oct 1, 2026, 02:26 PM
- Added 6 contextual internal links
Version 10 · Jul 23, 2026, 03:04 PM