VideoVIBE - State Transition: leaderboard

The State Transition subdimension (structural and temporal coherence) 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 State Transition 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 February 2027. 13 models tracked.

Top models

#ModelScoreOverall rank
1Qwen 3.5 35B A3B75.54#482
2Gemini 3.1 Flash Lite74.46#364
3Qwen 3 VL 30B A3B (Thinking)67.93#712
4Gemini 3 Flash66.3#164
5Gemini 3.1 Pro (Preview)65.76#94
6MiMo-V2.565.22#354
7Gemini 2.5 Flash63.59#453
8GLM-4.6V58.7#701
9Qwen 3 VL 30B A3B Instruct58.15#871
10Qwen 3 VL 8B (Thinking)57.61#887
11Qwen 3 VL 8B Instruct44.57#981

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=videovibe-state-transition · How It Works · Data refreshed daily, snapshot 2026-11-13.