Benchmark
VBench 1.0 subject consistency
Scores are percentages.
- Models measured
- 9
- Scores recorded
- 9
- Sources
- 1
- IQ weight
- 0.10
Score distribution
- 96.6%
- Mean
- 97%
- Median
- 0.896%
- Std dev
- 95.8%
- 10th percentile
- 97.4%
- 90th percentile
- 94.6%
- Min
This benchmark has a weight of 0.10 for video-generation models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Video 01· VBench data docs | 97.5% |
| 2 | Kling V1.6· VBench data docs | 97.4% |
| 3 | Veo 3· VBench data docs | 97.4% |
| 4 | Hunyuan Video2025-05-22· VBench data docs | 97.2% |
| 5 | Mochi 1· VBench data docs | 97% |
| 6 | LTX Video5s 768×512· VBench data docs | 96.6% |
| 7 | Sora· VBench data docs | 96.2% |
| 8 | Vidu Q12025-04-17· VBench data docs | 95.8% |
| 9 | Vidu Q4· VBench data docs | 94.6% |
Data credits: VBench (Vchitect), MIT (licence)