Video understanding benchmark testing models on questions about video content — temporal reasoning, event sequencing, and visual comprehension across video frames.
This definition has not completed editorial and provenance review. Verify its metric, version, source, and evaluation configuration upstream before using it for a model-selection decision.
Specifically tests video understanding, which image benchmarks miss. Requires temporal reasoning.
Requires video processing capabilities. Fewer evaluated models than image-based benchmarks.
Higher is better. Accuracy percentage (0–100%).
Requires video processing capabilities.
| # | Model | Score |
|---|---|---|
| 1 | 94.0 | |
| 2 | 94.0 | |
| 3 | 93.9 | |
| 4 | 93.9 | |
| 5 | 93.5 | |
| 6 | 93.4 | |
| 7 | 93.4 | |
| 8 | 93.3 | |
| 9 | 90.5 | |
| 10 | 90.5 | |
| 11 | 90.5 | |
| 12 | 89.3 | |
| 13 | 89.0 | |
| 14 | 88.5 | |
| 15 | 88.5 | |
| 16 | 88.3 | |
| 17 | 88.2 | |
| 18 | 87.8 | |
| 19 | 87.7 | |
| 20 | 87.6 | |
| 21 | 87.6 | |
| 22 | 87.6 | |
| 23 | 87.6 | |
| 24 | 87.6 | |
| 25 | 87.5 | |
| 26 | 87.4 | |
| 27 | 87.3 | |
| 28 | 87.2 | |
| 29 | 87.2 | |
| 30 | 87.2 | |
| 31 | 87.1 | |
| 32 | 87.1 | |
| 33 | 87.1 | |
| 34 | 87.0 | |
| 35 | 87.0 | |
| 36 | 87.0 | |
| 37 | 87.0 | |
| 38 | 87.0 | |
| 39 | 87.0 | |
| 40 | 86.9 | |
| 41 | 86.9 | |
| 42 | 86.9 | |
| 43 | 86.9 | |
| 44 | 86.9 | |
| 45 | 86.9 | |
| 46 | 86.9 | |
| 47 | 86.9 | |
| 48 | 86.9 | |
| 49 | 86.8 | |
| 50 | 86.8 |