Tool Use on StableToolBench G1 Instruction
75.5SL ScoreTrace-Based
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Trace-BasedEvaluation Protocol=Trace-based2026.02 | 75.5 | 66.9 | |
| D1Evaluation Protocol=Trace-based2026.02 | 74.7 | 66.1 | |
| D2Evaluation Protocol=Trace-based2026.02 | 73.4 | 62.5 | |
| D0Evaluation Protocol=Trace-based2026.02 | 72.8 | 62.3 | |
| Play2PromptEvaluation Protocol=Trace-based2026.02 | 72 | 62.5 | |
| DRAFTEvaluation Protocol=Trace-based2026.02 | 70 | 58 |