Tool Use Evaluation on ToolSandbox
0.923SimilarityWISE-Flow
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| WISE-FlowModel=Qwen3-235B-A22B2026.01 | 0.923 | 0.9615 | 3.69 | 88 | — | |
| REMEMBERERModel=Claude 3.7 Sonnet2026.01 | 0.914 | 0.9728 | 2.47 | 95 | — | |
| AutoGuideModel=Qwen3-235B-A22B2026.01 | 0.911 | 0.9484 | 5.08 | 87 | — | |
| ReActModel=Claude 3.7 Sonnet2026.01 | 0.911 | 0.9603 | 3.72 | 92 | — | |
| AutoGuideModel=Claude 3.7 Sonnet2026.01 | 0.911 | 0.9699 | 2.9 | 91 | — | |
| WISE-FlowModel=Claude 3.7 Sonnet2026.01 | 0.91 | 0.9745 | 2.46 | 95 | — | |
| ReflexionModel=Claude 3.7 Sonnet2026.01 | 0.875 | 0.9462 | 5.42 | 87 | — | |
| REMEMBERERModel=Qwen3-235B-A22B2026.01 | 0.871 | 0.9134 | 8.51 | 78 | — | |
| LLM-onlyModel=Claude 3.7 Sonnet2026.01 | 0.871 | 0.9277 | 6.99 | 87 | — | |
| ReflexionModel=Qwen3-235B-A22B2026.01 | 0.866 | 0.8999 | 9.79 | 84 | — | |
| ReActModel=Qwen3-235B-A22B2026.01 | 0.858 | 0.9008 | 9.7 | 76 | — | |
| LLM-onlyModel=Qwen3-235B-A22B2026.01 | 0.856 | 0.8985 | 9.75 | 77 | — | |
| TRACE2026.04 | 0.552 | — | — | — | 26 | |
| GEPA2026.04 | 0.52 | — | — | — | 22 | |
| GRPO on Target2026.04 | 0.519 | — | — | — | 22 | |
| Single Capability GRPO2026.04 | 0.514 | — | — | — | 22 | |
| AWM2026.04 | 0.504 | — | — | — | 20 | |
| ADP2026.04 | 0.422 | — | — | — | 19 | |
| BaseBackbone=Qwen3-30B-A3B2026.04 | 0.411 | — | — | — | 19 |