Galileo Tool Tasks - xLAM Single Tool Multiple Call Leaderboard — The Aggregate

Galileo Tool Tasks - xLAM Single Tool Multiple Call: leaderboard

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

Top models

#	Model	Score
1	GPT-4o Mini	98
2	Ministral-8B-Instruct-2410	98
3	Gemini 1.5 Pro	97
4	Mistral Small 3	97
5	Llama 3.1 8B Instruct	96
6	Gemini 2.0 Flash (001)	96
7	Claude 3.7 Sonnet (20250219)	95
8	GPT-4o (2024-11-20)	95
9	Mistral Large 2 (Nov) Instruct (2411)	93
10	Llama 3.3 70B Instruct	87
11	Gemini 1.5 Flash	87
12	Claude 3.5 Haiku (20241022)	79
13	Claude 3.5 Sonnet (20241022)	78
14	O3 Mini (2025-01-31)	72
15	O1 (2024-12-17)	71

Interactive version: theaggregate.ai/benchmark?slug=galileo-tool-tasks-xlam-single-tool-multiple-call · How It Works · Data refreshed daily, snapshot 2026-09-05.