How 20 AI image models render: “A man with a purple scarf holding a transparent umbrella, standing to the right of a woman in an orange raincoat who holds a paper cup, both facing the camera.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: every stated element present, correct count, correct placement. 135 community votes so far.
Core suite prompt · voting closed 2026-09-18
Checklist: man with purple scarf · transparent umbrella · woman in orange raincoat · paper cup in her hand · man to the RIGHT of the woman · both facing camera
20 text-to-image models rendered this prompt adherence prompt with the identical text at their default settings and a fixed seed. Qwen Image 3.0 Pro is the community favourite so far, winning 89% of its 9 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 4.3s.
What to look for: the right count of objects, the right spatial relationships, the stated colours and attributes, and nothing added.
- Qwen Image 3.0 Pro89%Alibaba10.3s · 36 cr Try
- Kling v3 Omni83%Kuaishou27.5s · 35 cr Try
- GPT Image 2.5 Flare82%OpenAI26.3s · 40 cr Try
- GPT Image 279%OpenAI · retired from voting58.2s · 60 cr Try
- 64%18.8s · 50 cr Try
- GPT Image 2.5 Sunburst59%OpenAI36.8s · 80 cr Try
- Seedream 5.0 Pro57%ByteDance32.5s · 49 cr Try
- Muse Image 1.053%Meta8.7s · 12 cr Try
- 4.3s · 80 cr Try
- Krea 2 Medium44%Krea31.8s · 40 cr Try
- ImagineArt 242%ImagineArt · retired from voting31.2s · 60 cr Try
- ImagineArt 1.5 Pro41%ImagineArt · retired from voting52.4s · 55 cr Try
- Recraft v4 Pro33%Recraft · retired from voting20.8s · 300 cr Try
- Nano Banana 2 Lite33%Google12.6s · 36 cr Try
- Qwen Image 3.027%Alibaba7.6s · 32 cr Try
- Recraft v423%Recraft · retired from voting13.0s · 45 cr Try
- Midjourney Niji 717%Midjourney · retired from voting104.8s · 160 cr Try
- Ideogram v410%Ideogram13.1s · 40 cr Try
- Krea 2 LargeKrea31.8s · 80 cr Try
FAQ
- Which AI model renders this prompt best?
- Qwen Image 3.0 Pro leads this prompt in the BudgetPixel AI Image Arena, winning 89% of its 9 blind head-to-head votes. Vote on the arena page to add your judgement.
- Were the prompts adjusted per model?
- No. Every model received the exact same prompt, unmodified, at its default settings, with no negative prompt or style suffix. Some models rewrite prompts internally, and that is part of what is being compared.
- Can I run this prompt myself?
- Yes. Each tile’s “Try” link opens the BudgetPixel workshop with this prompt and that model preselected.