Agentic Tool Calling on MAVEN-Bench
71Accuracy
MAVEN
Updated 2d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.05 | 71 | 95.1 | 67.4 | 18.2 | 9.5 | |
2026.05 | 70 | 94.2 | 67 | 17.7 | 9.5 | |
2026.05 | 57 | 92.4 | 65.5 | 17.7 | 9.3 | |
2026.05 | 55 | 88.2 | 63 | 16.8 | 8.4 | |
2026.05 | 48 | 88.9 | 62.6 | 17.5 | 8.7 | |
2026.05 | 43 | 88 | 62.3 | 16.9 | 8.7 | |
2026.05 | 38 | 76.7 | 53.6 | 15.7 | 7.5 | |
2026.05 | 32 | 61.1 | 43.3 | 12.6 | 5.2 |