Tool Use on StableToolBench G2 Category
71SLTrace-Based
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Trace-BasedEvaluation Protocol=Trace-based2026.02 | 71 | 50.7 | |
| D2Evaluation Protocol=Trace-based2026.02 | 70.7 | 47.1 | |
| D0Evaluation Protocol=Trace-based2026.02 | 68.4 | 42.4 | |
| D1Evaluation Protocol=Trace-based2026.02 | 67.9 | 43.1 | |
| Play2PromptEvaluation Protocol=Trace-based2026.02 | 66.7 | 42.4 | |
| DRAFTEvaluation Protocol=Trace-based2026.02 | 66.3 | 45.3 |