Multimodal Perception on MMMU
55.44ScoreLLaVA-OV
Evaluation Results
| Method | Links | |
|---|---|---|
| LLaVA-OVScale=1.5-8B2026.06 | 55.44 | |
| Qwen2.5-VLScale=7B2026.06 | 54.9 | |
| MOSS-Video-PreviewSFT Mode=real-time2026.06 | 47.36 | |
| MOSS-Video-PreviewSFT Mode=offline2026.06 | 45.26 | |
| Llama-3.2 11B-VisionCheckpoint=base2026.06 | 41.7 |