UI2App - Visual Fidelity: leaderboard
UI2App (Hong Kong University of Science and Technology, 2026) gives a vision-language model uncaptioned screenshots of a real multi-route web application, 45 sets and 327 screenshots in all, and asks it to generate a runnable React application. The Visual Fidelity Score (0-100, higher is better) compares rendered page blocks with the screenshots on size, text, position and colour; applications that never build score zero.
Metric: Visual Fidelity Score (0-100), the block-level size, text, position and colour similarity of each rendered route to its screenshot; 45 screenshot sets (327 screenshots) of real multi-route web applications; the model must emit a runnable React application from the screenshots alone; apps that fail to build after three self-debug rounds score 0. Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 78.1 | #94 |
| 2 | Claude Sonnet 4.6 | 75.7 | #161 |
| 3 | Kimi K2.5 (Thinking) | 65.3 | #256 (Kimi K2.5) |
| 4 | GPT-5.4 | 65 | #123 |
| 5 | Qwen 3.5 397B A17B | 56 | #246 |
| 6 | GLM-4.6V | 22.6 | #701 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=ui2app-visual-fidelity · How It Works · Data refreshed daily, snapshot 2026-11-13.