Human Creativity Benchmark - Ad Images (Refinement) - Usability: leaderboard
Human Creativity Benchmark (Hopkins et al., 2026): professional creatives rate anonymized advertising images at the Refinement stage, where a near-final design needs targeted edits. The score is the mean 1-5 rating of how well the output would work in the context of the prompt and campaign, over the evaluators and prompts of this domain and stage; higher is better.
Metric: Mean expert Likert rating (1-5) of usability in the context of the prompt and campaign; advertising images at the Refinement stage, rated by professional creatives. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seedream 4.5 | 4.23 | |
| 2 | GPT Image 1.5 | 3.73 | |
| 3 | Flux 2 Pro | 3.67 | |
| 4 | Nano Banana Pro (Gemini 3 Pro Image Preview) | 3.53 |
Interactive version: theaggregate.ai/benchmark?slug=human-creativity-benchmark-ad-images-refinement-usability · How It Works · Data refreshed daily, snapshot 2026-11-12.