Multimodal Perception Evaluation on MME
189.47Existence ScoreDAID
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DAIDBackbone=LLaVA-NeXT2026.04 | 189.47 | — | 133.12 | 156.29 | 165.52 | 644.4 | |
| VCDBackbone=LLaVA-NeXT2026.04 | 187.2 | — | 140.16 | 139.91 | 157.41 | 624.68 | |
| DAIDBackbone=LLaVA-1.5-7B2026.04 | 186.14 | — | 130.7 | 154.46 | 162.38 | 633.68 | |
| SIDBackbone=LLaVA-NeXT2026.04 | 185.03 | — | 134.96 | 137.23 | 161.74 | 618.96 | |
| VCDBackbone=LLaVA-1.5-7B2026.04 | 184.66 | — | 137.33 | 128.67 | 153 | 603.66 | |
| EAZYBackbone=LLaVA-NeXT2026.04 | 184.55 | — | 134.18 | 133.52 | 158.89 | 611.14 | |
| DoLaBackbone=LLaVA-NeXT2026.04 | 184.45 | — | 133.07 | 126.72 | 160.38 | 604.62 | |
| OPERABackbone=LLaVA-NeXT2026.04 | 183.98 | — | 135.85 | 118 | 131.72 | 569.55 | |
| SIDBackbone=LLaVA-1.5-7B2026.04 | 183.9 | — | 132.2 | 127.8 | 155.9 | 599.8 | |
| EAZYBackbone=LLaVA-1.5-7B2026.04 | 182.29 | — | 131.73 | 126.93 | 155.21 | 596.16 | |
| OPERABackbone=LLaVA-1.5-7B2026.04 | 180.67 | — | 133.33 | 111.67 | 123.33 | 549 | |
| Beam SearchBackbone=LLaVA-NeXT2026.04 | 180.27 | — | 128.9 | 120.34 | 155.29 | 584.8 | |
| DoLaBackbone=LLaVA-1.5-7B2026.04 | 180.1 | — | 127.4 | 119.3 | 154.6 | 581.4 | |
| GreedyBackbone=LLaVA-NeXT2026.04 | 179.78 | — | 127.4 | 121.92 | 151.82 | 580.92 | |
| Beam SearchBackbone=LLaVA-1.5-7B2026.04 | 175.67 | — | 124.67 | 114 | 151 | 565.34 | |
| GreedyBackbone=LLaVA-1.5-7B2026.04 | 173.45 | — | 122.72 | 113.39 | 149.92 | 559.48 | |
| AM-RADIO-HResolution=512², Tokens per image=256, LLM=Qwen2-7B-Instruct, Training Data=ShareGPT4v and VFLAN2024.12 | — | 1,516.5 | — | — | — | — | |
| DINOSAURv2temperature=0.02, number of seeds=52026.02 | — | 1,123.47 | — | — | — | — | |
| DINOv2temperature=0.02, number of seeds=52026.02 | — | 1,283.96 | — | — | — | — | |
| FT-DINOSAURtemperature=0.02, number of seeds=52026.02 | — | 1,016.15 | — | — | — | — | |
| mFRESAtemperature=0.02, number of seeds=52026.02 | — | 1,184.4 | — | — | — | — | |
| OpenAI-CLIPResolution=336², Tokens per image=144, LLM=Qwen2-7B-Instruct, Training Data=ShareGPT4v and VFLAN2024.12 | — | 1,646.9 | — | — | — | — | |
| RADIOv2.5-HResolution=768², Tokens per image=196, LLM=Qwen2-7B-Instruct, Training Data=ShareGPT4v and VFLAN2024.12 | — | 1,613.5 | — | — | — | — | |
| RADIOv2.5-HResolution=768², Tokens per image=512, LLM=Qwen2-7B-Instruct, Training Data=ShareGPT4v and VFLAN2024.12 | — | 1,587.6 | — | — | — | — | |
| SigLIP-SO400MResolution=384², Tokens per image=196, LLM=Qwen2-7B-Instruct, Training Data=ShareGPT4v and VFLAN2024.12 | — | 1,605.2 | — | — | — | — | |
| Slot Diffusiontemperature=0.02, number of seeds=52026.02 | — | 1,090.75 | — | — | — | — | |
| SPOTtemperature=0.02, number of seeds=52026.02 | — | 1,066.04 | — | — | — | — | |
| StableLSDtemperature=0.02, number of seeds=52026.02 | — | 1,126.06 | — | — | — | — |