Tool Usage on BFCL Multi-Parallel v1
90.5Accuracy
Qwen3-4B-Thinking
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2025.10 | 90.5 | - | |
2025.10 | 90.5 | -28.2 | |
2025.10 | 90 | -7.3 | |
2025.10 | 88.5 | -53.5 | |
2025.10 | 85 | -14.2 | |
2025.10 | 85 | - | |
2025.10 | 84 | 1.2 | |
2025.10 | 84 | 1.8 | |
2025.10 | 83.5 | -12 | |
2025.10 | 83.5 | - | |
2025.10 | 83 | -100 | |
2025.10 | 83 | - | |
2025.10 | 82.5 | -5 | |
2025.10 | 77.5 | - | |
2025.10 | 77.5 | -6.4 | |
2025.10 | 75.5 | -13.2 | |
2025.10 | 42.5 | -33.4 | |
2025.10 | 42.5 | -6.9 | |
2025.10 | 38 | - | |
2025.10 | 37 | -11.6 |