Reasoning Quality Evaluation on 120-tool benchmark 500 tasks simulated
4.43Mean Score
Tool Attention
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.04 | 4.43 | 0.62 | 87.6 | |
2026.04 | 4.02 | 0.77 | 74.1 | |
2026.04 | 3.89 | 0.81 | 68.7 | |
2026.04 | 3.35 | 0.98 | 48 | |
2026.04 | 3.21 | 1.04 | 43.2 |