Skip to content

Benchmark

VBench 1.0

Scores are percentages.

Models measured
9
Scores recorded
9
Sources
1
IQ weight
0.70

Score distribution

83.2%
Mean
83.4%
Median
2.2%
Std dev
80.1%
10th percentile
85.1%
90th percentile
80%
Min
80%87.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.70 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 docs87.4%
2Veo 3· VBench data docs85.1%
3Sora· VBench data docs84.3%
4Hunyuan Video2025-05-22· VBench data docs83.4%
5Video 01· VBench data docs83.4%
6Kling V1.6· VBench data docs83.4%
7Vidu Q4· VBench data docs81.9%
8Mochi 1· VBench data docs80.1%
9LTX Video5s 768×512· VBench data docs80%

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