ProductWebGen - Webpage Design Quality (Qwen-Image-Edit Pipeline): leaderboard
Product webpage generation (ProductWebGen, Liu et al., 2026): 500 requests in 13 product categories, each a source product image with layout, style, content and visual-consistency instructions, answered with renderable HTML and showcase images. An LLM writes the page and image descriptions and Qwen-Image-Edit draws the images; GLM-4.5V rates the rendered page for visual hierarchy, layout, color and aesthetics from 0 to 10, higher is better.
Metric: Webpage design quality (0-10; GLM-4.5V rating of the rendered page for visual hierarchy, layout, color and aesthetics), LLM page with Qwen-Image-Edit images, 500 samples. Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 7.95 | #452 |
| 2 | Grok 4 | 7.93 | #306 |
| 3 | Claude Sonnet 4 | 7.93 | #397 |
| 4 | GPT-4o | 7.72 | #741 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=productwebgen-webpage-design-quality-qwen-image-edit-pipeline · How It Works · Data refreshed daily, snapshot 2026-11-12.