Human Creativity Benchmark - Desktop Apps (Ideation) - Visual Appeal: leaderboard
Metric: Mean expert Likert rating (1-5) of visual interest, cohesion and polish; desktop-app interfaces written as code at the Ideation stage, rated by professional creatives. Source: arxiv.org. Saturation forecast: Around March 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 3.4 |
| 2 | Gemini 3.1 Pro (Preview) | 3.11 |
| 3 | GPT-5.3 Codex | 3.09 |
| 4 | Qwen 3.5 397B A17B | 2.71 |
Interactive version: theaggregate.ai/benchmark?slug=human-creativity-benchmark-desktop-apps-ideation-visual-appeal · How It Works · Data refreshed daily, snapshot 2026-09-26.