Qwen-Image-Bench - Human Expert Rating: leaderboard
Qwen-Image-Bench (2026) scores text-to-image models on 1,000 creator-designed prompts in five pillars: quality, aesthetics, alignment, real-world fidelity and creative generation. This board is the overall rating from professional annotators on a 1 to 10 scale; higher is better.
Metric: Overall professional annotator rating on a 1 to 10 scale over 1,000 prompts per pillar, averaged; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 18 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT Image 2 | 9.1 | |
| 2 | Nano Banana 2 | 8.71 | |
| 3 | Nano Banana Pro | 8.63 | |
| 4 | GPT Image 1.5 | 7.43 | |
| 5 | Qwen Image 2.0 Pro | 6.99 | |
| 6 | Seedream 5.0 | 6.44 | |
| 7 | Seedream 4.5 | 5.5 | |
| 8 | Qwen-Image-2512 | 5.45 | |
| 9 | GPT Image 1 | 5.25 | |
| 10 | Seedream 4.0 | 4.8 | |
| 11 | Imagen 4.0 Ultra | 4.72 | |
| 12 | Imagen 4.0 | 4.61 | |
| 13 | FLUX 2 Max | 4.49 | |
| 14 | FLUX 2 Pro | 4.33 | |
| 15 | HunyuanImage 3.0 | 3.37 |
Interactive version: theaggregate.ai/benchmark?slug=qwen-image-bench-human-expert-rating · How It Works · Data refreshed daily, snapshot 2026-11-22.