Galileo Tool Tasks - BFCL v3 Multi-Turn Missing Parameter: leaderboard
Tool-use evaluation from Galileo covering BFCL v3 cases where required function parameters are missing and must be requested or handled correctly.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.7 Sonnet (20250219) | 97 | #201 |
| 2 | O3 Mini (2025-01-31) | 93 | #240 |
| 3 | Gemini 2.0 Flash (001) | 92 | #360 |
| 4 | Gemini 1.5 Flash | 91 | #618 |
| 5 | O1 (2024-12-17) | 91 | #149 |
| 6 | Gemini 1.5 Pro | 90 | #445 |
| 7 | GPT-4o (2024-11-20) | 88 | #381 |
| 8 | Qwen 2.5 72B Instruct | 86 | #454 |
| 9 | Claude 3.5 Sonnet (20241022) | 85 | #291 |
| 10 | GPT-4o Mini | 83 | #599 |
| 11 | Claude 3.5 Haiku (20241022) | 79 | #588 |
| 12 | Mistral Small 3 | 76 | #627 |
| 13 | Mistral Large 2 (Nov) Instruct (2411) | 75 | #465 |
| 14 | Llama 3.1 8B Instruct | 66 | #1035 |
| 15 | Ministral-8B-Instruct-2410 | 66 | #1094 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=galileo-tool-tasks-bfcl-v3-multi-turn-missing-parameter · How It Works · Data refreshed daily, snapshot 2026-10-20.