Agentic Tool Use on Multi-step Tool-Calling Tasks (Model Aggregate Test)
99Success Rate
Gold CMTF
Updated 1mo ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.06 | 99 | 0.01 | 0 | 1 | 2,532 | |
2026.06 | 98 | 0.02 | 0 | 1 | 2,528 | |
2026.06 | 77.5 | 1.461 | 2.7 | 100 | 26,172 | |
2026.06 | 62 | 2.265 | 3.2 | 5 | 4,482 | |
2026.06 | 62 | 2.196 | 0 | 5.733 | 4,648 |