Coding on HumanEval++
81.71AccuracyGeneral-Reasoner
Evaluation Results
| Method | Links | |
|---|---|---|
| General-ReasonerGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 81.71 | |
| TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 81.71 | |
| RAMGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 81.71 | |
| DARE + TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 79.88 | |
| RESMERGEGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 78.66 | |
| TSV-MergeGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 78.05 | |
| Qwen3-4BGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 77.44 | |
| SpiralGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 76.22 | |
| TAGroup=Qwen3-4B-Base, Category=Merged Models2026.06 | 76.22 | |
| GRPOGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.06 | 75 |