General Performance on Aggregated Benchmarks
49.76Overall AverageGeneral-Reasoner
Evaluation Results
| Method | Links | |
|---|---|---|
| General-ReasonerGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 49.76 | |
| RESMERGEGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 47.25 | |
| TAGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 45.46 | |
| TSV-MergeGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 45.07 | |
| GRPOGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 43.57 | |
| RAMGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 42.22 | |
| RESMERGEBackbone=Qwen2.5-7B-Base, configuration=SRC-A+LHC2026.06 | 41.08 | |
| ReasonerBackbone=Qwen2.5-7B-Base2026.06 | 41.05 | |
| TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 40.93 | |
| ISO-CBackbone=Qwen2.5-7B-Base2026.06 | 38.86 | |
| ISO-CTSBackbone=Qwen2.5-7B-Base2026.06 | 38.77 | |
| TABackbone=Qwen2.5-7B-Base2026.06 | 38.66 | |
| SimpleRLBackbone=Qwen2.5-7B-Base2026.06 | 38.57 | |
| SpiralGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 38.29 | |
| DARE + TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 37.35 | |
| TSV-MergeBackbone=Qwen2.5-7B-Base2026.06 | 35.39 | |
| Qwen2.5-7BBackbone=Qwen2.5-7B-Base2026.06 | 34.32 | |
| TIESBackbone=Qwen2.5-7B-Base2026.06 | 32.83 | |
| RAMBackbone=Qwen2.5-7B-Base2026.06 | 31.7 | |
| DARE + TIESBackbone=Qwen2.5-7B-Base2026.06 | 30.64 | |
| ZeroBackbone=Qwen2.5-7B-Base2026.06 | 28.95 | |
| Qwen3-4BGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 28.71 |