AtelierEval - Image Constraint Satisfaction (Skilled Prompting): leaderboard
Metric: Image checklist satisfaction (%; share of each task's expert constraint checklist realized in the generated image, best of four images per backend, mean over four text-to-image backends (Gemini 3 Pro Image, GPT-Image-1, Flux.1 Pro, SDXL) and the three task categories, GPT-5.4 AtelierJudge; structured prompting with chain-of-thought and self-verification). Source: arxiv.org. Saturation forecast: Around 2030. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 65.4 |
| 2 | Gemini 3 Pro (Preview) | 65.4 |
| 3 | Claude Sonnet 4.5 | 65.1 |
| 4 | GPT-4.1 | 63.9 |
| 5 | Qwen 3 VL 235B A22B | 63.1 |
| 6 | Gemini 2.0 Flash | 62.9 |
| 7 | GPT-4.1 Nano | 62.8 |
| 8 | Qwen 3 VL 8B | 57.2 |
Interactive version: theaggregate.ai/benchmark?slug=ateliereval-image-constraint-satisfaction-skilled-prompting · How It Works · Data refreshed daily, snapshot 2026-09-26.