Multimodal Comprehension on MMVet
58ScoreQwen2-VL
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2-VLModel Size=7B, Training Strategy=Original2026.02 | 58 | |
| CREM_GModel Size=7B, Training Strategy=Fine-tuned on ShareGPT-4V2026.02 | 56.8 | |
| CREMModel Size=7B, Training Strategy=Unified framework2026.02 | 56.7 | |
| Qwen2-VLModel Size=2B, Training Strategy=Original2026.02 | 45.7 | |
| CREMModel Size=2B, Training Strategy=Unified framework2026.02 | 45.1 | |
| CREM_GModel Size=2B, Training Strategy=Fine-tuned on ShareGPT-4V2026.02 | 44.5 | |
| CREM_RModel Size=7B, Training Strategy=Trained on MMEB2026.02 | 41.4 | |
| CREM_RModel Size=2B, Training Strategy=Trained on MMEB2026.02 | 17.3 |