Tool-use on BFCL Single-Turn
84.11OAAWPO
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AWPOParameter=4B2025.12 | 84.11 | 87.9 | 80.32 | |
| ToolACE-2-8BParameter=8B2025.12 | 82.54 | 87.87 | 77.2 | |
| Qwen3-235B-A22BParameter=235B2025.12 | 82.46 | 87.9 | 77.03 | |
| BitAgent-8BParameter=8B2025.12 | 81.71 | 87.33 | 76.09 | |
| watt-tool-8BParameter=8B2025.12 | 81.71 | 87.54 | 75.87 | |
| Moonshotai-Kimi-K2-InstParameter=1000B2025.12 | 80.8 | 84.02 | 77.57 | |
| Grok-4-0709Parameter=/2025.12 | 79.8 | 85.21 | 74.39 | |
| DeepSeek-R1-0528Parameter=671B2025.12 | 78.22 | 75.73 | 80.9 | |
| Gemini-2.5-ProParameter=/2025.12 | 74.5 | 85.04 | 63.95 | |
| GPT-5-2025-08-07Parameter=/2025.12 | 65.59 | 72.92 | 58.25 |