Video Understanding on VideoMME (Accuracy Delta Reporting)
65.6Accuracy (Base)AKS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AKSModel=InternVL-3 [47], Size=8B, Frame=322026.02 | 65.6 | 66.3 | 0.7 | |
| MDP3Model=InternVL-3 [47], Size=8B, Frame=322026.02 | 65.6 | 66.8 | 1.2 | |
| A.I.R.Model=InternVL-3 [47], Size=8B, Frame=≤32§2026.02 | 65.6 | 68.2 | 2.6 | |
| WFS-SBModel=InternVL-3 [47], Size=8B, Frame=322026.02 | 65.6 | 67.4 | 1.8 | |
| AKSModel=Qwen2.5-VL [2], Size=7B, Frame=322026.02 | 61.2 | 64 | 2.8 | |
| WFS-SBModel=Qwen2.5-VL [2], Size=7B, Frame=322026.02 | 61.2 | 64.4 | 3.2 | |
| MDP3Model=Qwen2.5-VL [2], Size=7B, Frame=322026.02 | 60.8 | 63.8 | 3 | |
| A.I.R.Model=Qwen2.5-VL [2], Size=7B, Frame=≤32§2026.02 | 60.8 | 65 | 4.2 | |
| AKSModel=LLaVA-Video [44], Size=7B, Frame=82026.02 | 56.2 | 60.1 | 3.9 | |
| WFS-SBModel=LLaVA-Video [44], Size=7B, Frame=82026.02 | 56.2 | 61.7 | 5.5 | |
| BOLTModel=LLaVA-Video [44], Size=7B, Frame=82026.02 | 56 | 58.6 | 2.6 | |
| KFCModel=LLaVA-Video [44], Size=7B, Frame=82026.02 | 55.9 | 57.6 | 1.7 | |
| FrameOracleModel=LLaVA-Video [44], Size=7B, Frame=82026.02 | 55.9 | 58.9 | 3 | |
| KFCModel=Qwen2-VL-7B [34], Size=7B, Frame=82026.02 | 55 | 56.7 | 1.7 | |
| AKSModel=LLaVA-OV [10], Size=7B, Frame=82026.02 | 54.1 | 58.2 | 4.1 | |
| WFS-SBModel=LLaVA-OV [10], Size=7B, Frame=82026.02 | 54.1 | 59.3 | 5.2 | |
| BOLTModel=LLaVA-OV [10], Size=7B, Frame=82026.02 | 53.8 | 56.1 | 2.3 | |
| FrameOracleModel=LLaVA-OV [10], Size=7B, Frame=82026.02 | 53.8 | 57.5 | 3.7 | |
| Q-FrameModel=Qwen2-VL-7B [34], Size=7B, Frame=8 (4+8+32)∗2026.02 | 53.7 | 58.3 | 4.6 | |
| Frame-VoyagerModel=LLaVA-OV [10], Size=7B, Frame=82026.02 | 53.3 | 57.5 | 4.2 | |
| KFCModel=LLaVA-OV [10], Size=7B, Frame=82026.02 | 53.3 | 55.4 | 2.1 | |
| WFS-SBModel=Qwen2-VL-7B [34], Size=7B, Frame=82026.02 | 52 | 56.4 | 4.4 |