Tool Use on BFCL (v4, v3, v4+H)
77BFCL v4 Score
Vanilla
Updated 18d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.07 | 77 | 42 | 85.5 | |
2026.07 | 73 | 41.5 | 91.5 | |
2026.07 | 73 | 45 | 83.5 | |
2026.07 | 70.5 | 42.5 | 89 | |
2026.07 | 70.5 | 43.5 | 85 | |
2026.07 | 70.5 | 39.5 | 87.5 | |
2026.07 | 70 | 39 | 75 | |
2026.07 | 70 | 41 | 87.5 | |
2026.07 | 67 | 44 | 85.5 | |
2026.07 | 64 | 36 | 90 | |
2026.07 | 62 | 42 | 73 | |
2026.07 | 61 | 30 | 69.5 | |
2026.07 | 57.5 | 31 | 74 | |
2026.07 | 57 | 30.5 | 75.5 | |
2026.07 | 56 | 29 | 68 | |
2026.07 | 46.5 | 28 | 61.5 | |
2026.07 | 38.5 | 22 | 56 | |
2026.07 | 26.5 | 19.5 | 41.5 | |
2026.07 | 15.5 | 1 | 28 | |
2026.07 | 2.5 | 1.5 | 5.5 |