Skip to content

Benchmark

VBench 1.0 color

Scores are percentages.

Models measured
9
Scores recorded
9
Sources
1
IQ weight
0.60

Score distribution

85.3%
Mean
82.5%
Median
6.1%
Std dev
80.1%
10th percentile
90.4%
90th percentile
79.7%
Min
79.7%99%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.60 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

#ModelScore
1Vidu Q12025-04-17· VBench data docs99%
2Video 01· VBench data docs90.4%
3Hunyuan Video2025-05-22· VBench data docs89.8%
4Vidu Q4· VBench data docs83.2%
5Veo 3· VBench data docs82.5%
6LTX Video5s 768×512· VBench data docs81.5%
7Kling V1.6· VBench data docs81.3%
8Sora· VBench data docs80.1%
9Mochi 1· VBench data docs79.7%

Data credits: VBench (Vchitect), MIT (licence)