VideoVIBE - Layout Integrity: leaderboard
Metric: Failure-diagnosis accuracy (%; questions of the Layout Integrity subdimension of Structural and Temporal Coherence; 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. Saturation forecast: Around May 2028. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 30B A3B (Thinking) | 82.8 |
| 2 | Qwen 3 VL 8B (Thinking) | 78.98 |
| 3 | Gemini 3.1 Pro (Preview) | 77.07 |
| 4 | Gemini 3.1 Flash Lite | 71.97 |
| 5 | Qwen 3 VL 30B A3B Instruct | 71.97 |
| 6 | Gemini 2.5 Flash | 68.15 |
| 7 | GLM-4.6V | 66.88 |
| 8 | Qwen 3.5 35B A3B | 63.69 |
| 9 | MiMo-V2.5 | 61.15 |
| 10 | Qwen 3 VL 8B Instruct | 50.32 |
| 11 | Gemini 3 Flash | 41.4 |
Interactive version: theaggregate.ai/benchmark?slug=videovibe-layout-integrity · How It Works · Data refreshed daily, snapshot 2026-09-29.