Multimodal Perception on MMBench-EN (dev)
87.76ScoreQwen2.5-VL
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-VLScale=7B2026.06 | 87.76 | |
| LLaVA-OVScale=1.5-8B2026.06 | 84.14 | |
| MOSS-Video-PreviewSFT Mode=real-time2026.06 | 83.97 | |
| MOSS-Video-PreviewSFT Mode=offline2026.06 | 83.09 | |
| Llama-3.2 11B-VisionCheckpoint=base2026.06 | 72.76 |