Multimodal Evaluation on MME Real-World
38.9ScoreCapRL-Image-5M
Evaluation results
| Method | Links | |
|---|---|---|
| CapRL-Image-5MModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 38.9 | |
| CapRL-Image-5MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 34.9 | |
| DenseFusion-1MModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 34.1 | |
| CapRL-Image-1MModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 34.1 | |
| ShareGPT4V-1MModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 33.2 | |
| CapRL-Image-1MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 32.2 | |
| ShareGPT4V-1MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 31.8 | |
| DenseFusion-1MModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 31.1 | |
| CapRL-Image-5MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 30.9 | |
| DenseFusion-1MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 30.7 | |
| VanillaModel Configuration=InternLM2.5-7B + CLIP-ViT-L2026.06 | 30.7 | |
| VanillaModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 30.5 | |
| CapRL-Image-1MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 30.3 | |
| ShareGPT4V-1MModel Configuration=Qwen2.5-3B + Qwen2.5-ViT2026.06 | 29.8 | |
| VanillaModel Configuration=Qwen2.5-7B + Qwen2.5-ViT2026.06 | 29.4 |