| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MSRVTT-QA | Flash-VStream | Accuracy72.4 | 513 | 1mo ago | |
| ActivityNet-QA | Qwen3-VL + CurEvo | Accuracy71.02 | 438 | 1mo ago | |
| MSVD-QA | Flash-VStream | Accuracy80.3 | 401 | 1mo ago | |
| MSRVTT-QA (test) | CLIPBERT | Accuracy88.2 | 376 | 4mo ago | |
| ActivityNet-QA (test) | Accuracy82.78 | 288 | 4mo ago | ||
| MSVD-QA (test) | Video-QTR | Accuracy87.8 | 279 | 4mo ago | |
| EgoSchema (Full) | TASKER | Accuracy77.3 | 256 | 25d ago | |
| VideoMME | Gemini 2.5 Pro | Accuracy85.1 | 254 | 25d ago | |
| LongVideoBench | Gemini-2.5-Pro | Accuracy77.6 | 224 | 18d ago | |
| MLVU | VideoChat-A1 | Accuracy76.2 | 213 | 1mo ago | |
| NExT-QA (test) | Accuracy86.3 | 204 | 4mo ago | ||
| EgoSchema | A4VL | Accuracy82.2 | 194 | 25d ago | |
| NExT-QA (val) | Overall Acc88.4 | 176 | 4mo ago | ||
| MSVD | LVLM | Accuracy79.5 | 169 | 1mo ago | |
| VideoMMMU | Gemini 2.5 Pro | Accuracy74.9 | 166 | 1mo ago | |
| TGIF-QA | HiTeA | Accuracy97.2 | 161 | 1mo ago | |
| EgoSchema subset | VideoHV-Agent | Accuracy81 | 124 | 2mo ago | |
| NEXT-QA | LinVT-Qwen2-VL | Overall Accuracy85.5 | 122 | 18d ago | |
| MSRVTT | LVLM | Accuracy66.7 | 117 | 1mo ago | |
| NExT-QA Multi-choice | VideoLLaMA3-7B | Accuracy84.5 | 114 | 4mo ago | |
| LongVideoBench (val) | HAVEN | Accuracy83 | 113 | 1mo ago | |
| LVBench | HAVEN | Accuracy84.1 | 108 | 3mo ago | |
| NextQA | Penguin-VL | Accuracy85.4 | 92 | 1mo ago | |
| EgoSchema (test) | QwenVL2 | Accuracy77.9 | 90 | 3mo ago | |
| MVBench | Gemini 1.5 Pro | Accuracy81.3 | 90 | 4mo ago |