Tool Use on StableToolBench G3 Instruction
66.3SL ScorePlay2Prompt
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Play2PromptEvaluation Protocol=Trace-based2026.02 | 66.3 | 45.1 | |
| DRAFTEvaluation Protocol=Trace-based2026.02 | 63.2 | 42.6 | |
| Trace-BasedEvaluation Protocol=Trace-based2026.02 | 60.7 | 45.1 | |
| D2Evaluation Protocol=Trace-based2026.02 | 57.4 | 38.3 | |
| D0Evaluation Protocol=Trace-based2026.02 | 50.7 | 24.6 | |
| D1Evaluation Protocol=Trace-based2026.02 | 45.2 | 25.4 |