Qwen-Image-Bench (English) - Creative Generation: 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 creative generation (expressing authentic creative intent and style) pillar score from the judge model on the English prompts, 0 to 100; higher is better.

Metric: Creative generation (expressing authentic creative intent and style) pillar Q-Judger score (0-100), the unified judge model fine-tuned from Qwen3.6-27B on expert annotations, aggregated bottom-up from 56 third-level facets, over the 1,000 creator-designed English prompts; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 18 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT Image 275.34
2GPT Image 1.567.06
3Nano Banana Pro64.54
4Nano Banana 264.49
5Seedream 5.063.64
6Qwen Image 2.0 Pro63.59
7Seedream 4.559.82
8FLUX 2 Max57.79
9Seedream 4.057.64
10FLUX 2 Pro57.36
11GPT Image 157.29
12Imagen 4.0 Ultra51.32
13Imagen 4.048.56
14HunyuanImage 3.048.33
15Qwen-Image-251247.75

Interactive version: theaggregate.ai/benchmark?slug=qwen-image-bench-english-creative-generation · How It Works · Data refreshed daily, snapshot 2026-11-21.