Gorilla API Bench (BFCL): leaderboard
Berkeley Function-Calling Leaderboard (Gorilla) — function/tool-calling accuracy including parallel and multi-turn calls, checked by AST matching and execution.
Metric: Overall Accuracy (%). Source: gorilla.cs.berkeley.edu. Saturation forecast: Around January 2027. 109 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Medium 3 | 37.69 | #500 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=gorilla-api-bench-bfcl · How It Works · Data refreshed daily, snapshot 2026-10-20.