List your company

Multimodal Reasoning on MMMU (offline) (Accuracy)

66.91AccuracySkeMex
41.471648.075854.6861.2842Jun 8, 2026

Evaluation results

MethodLinks
SkeMexBackbone=DeepSeek-V3.2, Setting=Offline2026.0666.91
CRITICBackbone=DeepSeek-V3.2, Setting=Offline2026.0663.31
DILUBackbone=DeepSeek-V3.2, Setting=Offline2026.0662.59
SkillWeaverBackbone=DeepSeek-V3.2, Setting=Offline2026.0662.59
GMBackbone=DeepSeek-V3.2, Setting=Offline2026.0661.87
SkeMexBackbone=Qwen3.6-Plus, Setting=Offline2026.0661.87
CFMBackbone=DeepSeek-V3.2, Setting=Offline2026.0661.15
EvolverBackbone=DeepSeek-V3.2, Setting=Offline2026.0660.43
GSEMBackbone=DeepSeek-V3.2, Setting=Offline2026.0660.43
VoyagerBackbone=DeepSeek-V3.2, Setting=Offline2026.0659.71
ReflexionBackbone=DeepSeek-V3.2, Setting=Offline2026.0658.99
MempBackbone=DeepSeek-V3.2, Setting=Offline2026.0658.99
Agent KBBackbone=DeepSeek-V3.2, Setting=Offline2026.0658.99
ReflexionBackbone=Qwen3.6-Plus, Setting=Offline2026.0658.99
SkillWeaverBackbone=Qwen3.6-Plus, Setting=Offline2026.0658.99
CoTBackbone=HuluMed-32B, Setting=Offline2026.0658.27
DCBackbone=DeepSeek-V3.2, Setting=Offline2026.0658.27
MobileEBackbone=DeepSeek-V3.2, Setting=Offline2026.0658.27
CRITICBackbone=Qwen3.6-Plus, Setting=Offline2026.0658.27
AWMBackbone=Qwen3.6-Plus, Setting=Offline2026.0658.27
CFMBackbone=Qwen3.6-Plus, Setting=Offline2026.0658.27
GSEMBackbone=Qwen3.6-Plus, Setting=Offline2026.0658.27
ExPeLBackbone=DeepSeek-V3.2, Setting=Offline2026.0657.55
ExPeLBackbone=Qwen3.6-Plus, Setting=Offline2026.0657.55
GMBackbone=Qwen3.6-Plus, Setting=Offline2026.0657.55
AWMBackbone=DeepSeek-V3.2, Setting=Offline2026.0656.83
MempBackbone=Qwen3.6-Plus, Setting=Offline2026.0655.4
MobileEBackbone=Qwen3.6-Plus, Setting=Offline2026.0655.4
Agent KBBackbone=Qwen3.6-Plus, Setting=Offline2026.0654.68
EvolverBackbone=Qwen3.6-Plus, Setting=Offline2026.0654.68
DCBackbone=Qwen3.6-Plus, Setting=Offline2026.0653.24
VoyagerBackbone=Qwen3.6-Plus, Setting=Offline2026.0652.52
DILUBackbone=Qwen3.6-Plus, Setting=Offline2026.0652.52
CoTBackbone=Lingshu-32B, Setting=Offline2026.0649.64
ReActBackbone=DeepSeek-V3.2, Setting=Offline2026.0646.04
ReActBackbone=Qwen3.6-Plus, Setting=Offline2026.0645.32
CoTBackbone=MedGemma-27B, Setting=Offline2026.0644.6
CoTBackbone=DeepSeek-V3.2, Setting=Offline2026.0642.45