Multimodal Understanding on CCBench 80% (test)
73.96AccuracyQwen2-VL-M3ID
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2-VL-M3IDBase Model=Qwen2-VL, Inference Strategy=M3ID2026.03 | 73.96 | |
| Qwen2-VL-TIEBase Model=Qwen2-VL, Inference Strategy=TIE2026.03 | 73.84 | |
| Qwen2-VL-VCDBase Model=Qwen2-VL, Inference Strategy=VCD2026.03 | 73.77 | |
| Qwen2-VL-SCI5Base Model=Qwen2-VL, Inference Strategy=SCI52026.03 | 73.59 | |
| Qwen2-VLBase Model=Qwen2-VL2026.03 | 73.22 | |
| LLaVA-NeXTBase Model=LLaVA-NeXT2026.03 | 47 | |
| LLaVA-NeXT-VCDBase Model=LLaVA-NeXT, Inference Strategy=VCD2026.03 | 46.63 | |
| LLaVA-NeXT-SCI5Base Model=LLaVA-NeXT, Inference Strategy=SCI52026.03 | 46.2 | |
| LLaVA-NeXT-M3IDBase Model=LLaVA-NeXT, Inference Strategy=M3ID2026.03 | 45.89 | |
| LLaVA-NeXT-TIEBase Model=LLaVA-NeXT, Inference Strategy=TIE2026.03 | 45.65 |