Tool Using on Non-Live Para
92AccuracyAMPO
Evaluation Results
| Method | Links | |
|---|---|---|
| AMPOModel Category=Base and Expert Models2026.06 | 92 | |
| Qwen2.5-7B-InstructModel Category=Base and Expert Models2026.06 | 91.5 | |
| TAModel Category=Merged Models2026.06 | 91 | |
| ISO-CTSModel Category=Merged Models2026.06 | 91 | |
| RESMERGEModel Category=Merged Models2026.06 | 91 | |
| ToolRLModel Category=Base and Expert Models2026.06 | 90.5 | |
| TSV-MergeModel Category=Merged Models2026.06 | 90.5 | |
| RAMModel Category=Merged Models2026.06 | 90.5 | |
| TIESModel Category=Merged Models2026.06 | 90 | |
| ISO-CModel Category=Merged Models2026.06 | 90 | |
| MemoryAgentModel Category=Base and Expert Models2026.06 | 89.5 | |
| DARE + TIESModel Category=Merged Models2026.06 | 89 | |
| General-ReasonerGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 78.5 | |
| TAGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 76.5 | |
| RESMERGEGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 74.5 | |
| GRPOGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 68.5 | |
| TSV-MergeGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 67.5 | |
| SpiralGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 57 | |
| Qwen3-4BGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 56.5 | |
| RAMGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 44.5 | |
| TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 43.5 | |
| DARE + TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 33.5 |