Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

Galileo Tool Tasks - BFCL v3 Multi-Turn Composite — leaderboard

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

Top models

#ModelScore
1Claude 3.7 Sonnet (20250219)96
2Gemini 2.0 Flash (001)94
3Gemini 1.5 Pro93
4O1 (2024-12-17)92
5O3 Mini (2025-01-31)91
6Gemini 1.5 Flash90
7GPT-4o (2024-11-20)90
8Qwen 2.5 72B Instruct87
9GPT-4o Mini85
10Claude 3.5 Sonnet (20241022)81
11Mistral Small 378
12Mistral Large 2 (Nov) Instruct (2411)76
13Claude 3.5 Haiku (20241022)72
14Ministral-8B-Instruct-241059
15Llama 3.3 70B Instruct50

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

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.