| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| VideoMMMU | Video-Zero | Accuracy68.33 | 141 | 18d ago | |
| VSI-Bench | LATENT-VC-9B | Accuracy61.9 | 101 | 18d ago | |
| Video-Holmes | Seed2.0 Pro | Accuracy67.4 | 89 | 23d ago | |
| Video-MMMU | Accuracy84.6 | 83 | 23d ago | ||
| Video-MME | Ola-7B | Accuracy68.4 | 73 | 18d ago | |
| MMVU | Accuracy75.8 | 71 | 18d ago | ||
| VideoMathQA | CoPD | Accuracy55.76 | 67 | 1mo ago | |
| LongVideoReason | Video-Zero | Accuracy73.1 | 61 | 1mo ago | |
| MVBench | MVBench Score84.3 | 56 | 18d ago | ||
| TempCompass | Score86.7 | 46 | 18d ago | ||
| VBVR-Bench Out-of-Domain | VBVR-Wan2.2 | Average Score61 | 39 | 1mo ago | |
| VBVR-Bench In-Domain | Average Score96 | 35 | 1mo ago | ||
| Video-Holmes | Score46.7 | 34 | 3mo ago | ||
| MMVU mc | Score82.6 | 32 | 1mo ago | ||
| VSI-Bench (test) | Video-ToC | Accuracy38.6 | 29 | 3mo ago | |
| SAGE-Bench 1.0 (test) | SAGE-Flash | Overall Score73.4 | 29 | 4mo ago | |
| VideoMathQA MCQ | DBTrimKV | Accuracy36.43 | 27 | 2mo ago | |
| VideoMMMU comprehension | DBTrimKV | Accuracy61.33 | 27 | 2mo ago | |
| VideoMMMU adaptation | DBTrimKV | Accuracy39 | 27 | 2mo ago | |
| Seed-Bench R1 | APPO | Average Answer Score50.5 | 26 | 4mo ago | |
| MMVU Multiple-choice (test) | GPT-4o | Accuracy75.4 | 25 | 2mo ago | |
| LVBench | Triage | LVBench Score43.3 | 24 | 4mo ago | |
| LongVideoBench | Triage | LongVideoBench Score59 | 24 | 1mo ago | |
| EgoSchema (test) | CLiViS (InternVL3) | Accuracy69.4 | 23 | 2mo ago | |
| Video-Holmes | VideoSeek | SR56.1 | 22 | 4mo ago |