Multimodal Reasoning on MMMU (offline) (Accuracy)
66.91AccuracySkeMex
Evaluation Results
| Method | Links | |
|---|---|---|
| SkeMexBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 66.91 | |
| CRITICBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 63.31 | |
| DILUBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 62.59 | |
| SkillWeaverBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 62.59 | |
| GMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 61.87 | |
| SkeMexBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 61.87 | |
| CFMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 61.15 | |
| EvolverBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 60.43 | |
| GSEMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 60.43 | |
| VoyagerBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 59.71 | |
| ReflexionBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 58.99 | |
| MempBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 58.99 | |
| Agent KBBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 58.99 | |
| ReflexionBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 58.99 | |
| SkillWeaverBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 58.99 | |
| CoTBackbone=HuluMed-32B, Setting=Offline2026.06 | 58.27 | |
| DCBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 58.27 | |
| MobileEBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 58.27 | |
| CRITICBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 58.27 | |
| AWMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 58.27 | |
| CFMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 58.27 | |
| GSEMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 58.27 | |
| ExPeLBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 57.55 | |
| ExPeLBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 57.55 | |
| GMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 57.55 | |
| AWMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 56.83 | |
| MempBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 55.4 | |
| MobileEBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 55.4 | |
| Agent KBBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 54.68 | |
| EvolverBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 54.68 | |
| DCBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 53.24 | |
| VoyagerBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 52.52 | |
| DILUBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 52.52 | |
| CoTBackbone=Lingshu-32B, Setting=Offline2026.06 | 49.64 | |
| ReActBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 46.04 | |
| ReActBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 45.32 | |
| CoTBackbone=MedGemma-27B, Setting=Offline2026.06 | 44.6 | |
| CoTBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 42.45 |