Human Creativity Benchmark - Desktop Apps (Mockup) - Prompt Adherence: leaderboard
Human Creativity Benchmark (Hopkins et al., 2026): professional creatives rate anonymized desktop-app interfaces generated as code at the Mockup stage, where a chosen direction is developed into a coherent concept. The score is the mean 1-5 rating of how faithful the output is to the prompt, over the evaluators and prompts of this domain and stage; higher is better.
Metric: Mean expert Likert rating (1-5) of faithfulness to the prompt; desktop-app interfaces written as code at the Mockup stage, rated by professional creatives. Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 3.97 | #100 |
| 2 | GPT-5.3 Codex | 3.94 | #114 |
| 3 | Gemini 3.1 Pro (Preview) | 3.89 | #91 |
| 4 | Qwen 3.5 397B A17B | 3.63 | #242 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=human-creativity-benchmark-desktop-apps-mockup-prompt-adherence · How It Works · Data refreshed daily, snapshot 2026-11-12.