AnimationBench - Semantic Consistency: leaderboard
Metric: Semantic Consistency overall: mean of object, action, color and scene agreement with the prompt and conditioning image (the paper's Sem. Overall) (%), animation-style image-to-video generation from the AnimationBench close-set suite (fixed source images and prompts, about 30 videos per dimension per model), scored by Qwen3-VL-Max yes/no question sets aggregated per video; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Veo 3.1 | 89.31 |
| 2 | Seedance 2.0 | 89.05 |
| 3 | Kling 2.6 | 86.11 |
| 4 | Sora 2 Pro | 73.36 |
| 5 | Seedance 1.0 Pro | 69.86 |
| 6 | Wan2.2-I2V-A14B | 67.86 |
| 7 | FramePack | 33.33 |
| 8 | HunyuanVideo-1.5 | 24.2 |
Interactive version: theaggregate.ai/benchmark?slug=animationbench-semantic-consistency · How It Works · Data refreshed daily, snapshot 2026-10-07.