Galileo Tool Tasks - BFCL v3 Multi-Turn Long Context Leaderboard — The Aggregate

Galileo Tool Tasks - BFCL v3 Multi-Turn Long Context — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 19 models tracked.

Top models

#	Model	Score
1	Claude 3.7 Sonnet (20250219)	94
2	Gemini 1.5 Pro	91
3	Gemini 2.0 Flash (001)	91
4	O1 (2024-12-17)	88
5	Gemini 1.5 Flash	87
6	O3 Mini (2025-01-31)	85
7	GPT-4o Mini	83
8	GPT-4o (2024-11-20)	83
9	Mistral Small 3	80
10	Llama 3.3 70B Instruct	79
11	Qwen 2.5 72B Instruct	79
12	Ministral-8B-Instruct-2410	78
13	Mistral Large 2 (Nov) Instruct (2411)	77
14	Llama 3.1 8B Instruct	73
15	Claude 3.5 Haiku (20241022)	73

Interactive version: theaggregate.ai/benchmark?slug=galileo-tool-tasks-bfcl-v3-multi-turn-long-context · How the rankings work · Data refreshed daily, snapshot 2026-07-22.