Tool Usage on BFCL Parallel v2
87.5Accuracy
Qwen3-4B-Thinking
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2025.10 | 87.5 | - | |
2025.10 | 87.5 | -32.3 | |
2025.10 | 81.3 | -50.7 | |
2025.10 | 81.2 | - | |
2025.10 | 81.2 | - | |
2025.10 | 81.2 | -6.4 | |
2025.10 | 75 | -5.1 | |
2025.10 | 75 | 4.6 | |
2025.10 | 68.8 | - | |
2025.10 | 68.8 | -3.9 | |
2025.10 | 68.8 | -4.6 | |
2025.10 | 62.5 | -28.5 | |
2025.10 | 62.5 | -21.1 | |
2025.10 | 62.5 | - | |
2025.10 | 62.5 | -30.1 | |
2025.10 | 62.5 | -25.4 | |
2025.10 | 50 | -8.9 | |
2025.10 | 43.8 | - | |
2025.10 | 43.8 | -19.9 | |
2025.10 | 37.5 | -100 |