VideoVIBE - Motion Quality: leaderboard
The Motion Quality subdimension (visual and motion fidelity) of VideoVIBE (2026): video MLLMs diagnose which failure a recorded session of a one-shot generated website shows, from the video and page source. Accuracy in percent; higher is better.
Metric: Failure-diagnosis accuracy (%; questions of the Motion Quality subdimension of Visual and Motion Fidelity; zero-shot single-choice diagnosis from the interaction video and the page source code (Video+HTML), about 1.7K questions built from 6,338 verified failures of one-shot generated web applications recorded under human operation; free-form answers mapped deterministically to an option). Source: arxiv.org. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 30B A3B (Thinking) | 84.56 | #712 |
| 2 | Qwen 3 VL 8B (Thinking) | 80 | #887 |
| 3 | Qwen 3 VL 30B A3B Instruct | 75.79 | #871 |
| 4 | Gemini 2.5 Flash | 75.09 | #453 |
| 5 | Qwen 3.5 35B A3B | 69.47 | #482 |
| 6 | Qwen 3 VL 8B Instruct | 64.91 | #981 |
| 7 | GLM-4.6V | 63.86 | #701 |
| 8 | Gemini 3.1 Pro (Preview) | 60.35 | #94 |
| 9 | Gemini 3.1 Flash Lite | 59.65 | #364 |
| 10 | MiMo-V2.5 | 44.91 | #354 |
| 11 | Gemini 3 Flash | 35.09 | #164 |
Interactive version: theaggregate.ai/benchmark?slug=videovibe-motion-quality · How It Works · Data refreshed daily, snapshot 2026-11-13.