Visual Question Answering on MME
1,648.1AccuracyInternVL-2.0
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| InternVL-2.0Params (B)=8, Text Gen Arch=AR, Image Gen Arch=-2025.05 | 1,648.1 | — | — | — | |
| LLaVA-NextParams (B)=13, Text Gen Arch=AR, Image Gen Arch=-2025.05 | 1,575 | — | — | — | |
| QWEN-VLParams (B)=7, Text Gen Arch=AR, Image Gen Arch=-2025.05 | 1,487.5 | — | — | — | |
| BLIP-2Params (B)=13, Text Gen Arch=AR, Image Gen Arch=-2025.05 | 1,293.8 | — | — | — | |
| MudditParams (B)=1, Text Gen Arch=Discrete Diff., Image Gen Arch=Discrete Diff., Resolution=1024×10242025.05 | 1,139.2 | — | — | — | |
| D-DiTParams (B)=2, Text Gen Arch=Discrete Diff., Image Gen Arch=Diffusion, Resolution=512×5122025.05 | 1,124.7 | — | — | — | |
| MudditParams (B)=1, Text Gen Arch=Discrete Diff., Image Gen Arch=Discrete Diff., Resolution=512×5122025.05 | 1,107.4 | — | — | — | |
| Show-OParams (B)=1.3, Text Gen Arch=AR, Image Gen Arch=Discrete Diff., Resolution=512×5122025.05 | 1,097.2 | — | — | — | |
| Show-OParams (B)=1.3, Text Gen Arch=AR, Image Gen Arch=Discrete Diff., Resolution=256×2562025.05 | 1,014.9 | — | — | — | |
| D-DiTParams (B)=2, Text Gen Arch=Discrete Diff., Image Gen Arch=Diffusion, Resolution=256×2562025.05 | 897.5 | — | — | — | |
| EAGLEAgent=Qwen, GLM, InternVL2026.05 | 87.9 | — | — | — | |
| Debate(Vote)Agent=Qwen, GLM, InternVL2026.05 | 85.2 | — | — | — | |
| ReConcileAgent=Qwen, GLM, InternVL2026.05 | 84.9 | — | — | — | |
| Debate(Judge)Agent=Qwen, GLM, InternVL2026.05 | 84.8 | — | — | — | |
| DARTAgent=Qwen, GLM, InternVL2026.05 | 84.4 | — | — | — | |
| Self-ConsistencyAgent=Qwen2026.05 | 84.2 | — | — | — | |
| Zero-shot CoTAgent=Qwen2026.05 | 83.6 | — | — | — | |
| Zero-shot CoTAgent=GLM2026.05 | 78.5 | — | — | — | |
| Zero-shot CoTAgent=InternVL2026.05 | 78 | — | — | — | |
| Self-RefineAgent=Qwen2026.05 | 30.6 | — | — | — | |
| LLaVA-1.5 13BIntervention Strategy=No-intervention baseline2026.01 | — | 79.36 | 32.31 | 47.89 | |
| LLaVA-1.5 13BIntervention Strategy=Q4 system redistribution (prop)2026.01 | — | 79.32 | 27.69 | 54.93 |