Portrait & fashion
How 20 AI image models render: “Portrait of an adult woman in her 30s wearing a rust-colored velvet jacket over an ivory blouse, one hand resting beneath her chin, thoughtful expression, deep teal studio backdrop, soft side lighting and gentle cheek highlights.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: face and hands, fabric, believable expression. 190 community votes so far.
Fresh prompt from the daily stream · voting closed 2026-09-19
20 text-to-image models rendered this portrait & fashion prompt with the identical text at their default settings and a fixed seed. GPT Image 2.5 Sunburst is the community favourite so far, winning 77% of its 13 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 4.1s.
What to look for: correct anatomy, natural expressions, convincing fabric drape, and hands with the right number of fingers.
- GPT Image 2.5 Sunburst77%OpenAI25.5s · 28 cr Try
- Muse Image 1.070%Meta9.9s · 12 cr Try
- GPT Image 267%OpenAI · retired from voting45.2s · 60 cr Try
- Qwen Image 3.0 Pro63%Alibaba9.1s · 36 cr Try
- ImagineArt 262%ImagineArt · retired from voting31.7s · 60 cr Try
- Kling v3 Omni57%Kuaishou27.1s · 35 cr Try
- 50%17.5s · 50 cr Try
- Ideogram v450%Ideogram10.6s · 40 cr Try
- Nano Banana 2 Lite48%Google10.2s · 36 cr Try
- GPT Image 2.5 Flare45%OpenAI27.2s · 28 cr Try
- Qwen Image 3.044%Alibaba12.9s · 32 cr Try
- ImagineArt 1.5 Pro43%ImagineArt · retired from voting48.7s · 55 cr Try
- 4.1s · 80 cr Try
- Seedream 5.0 Pro40%ByteDance31.3s · 49 cr Try
- Krea 2 Large39%Krea35.9s · 80 cr Try
- Recraft v4 Pro37%Recraft · retired from voting20.4s · 300 cr Try
- Krea 2 Medium35%Krea23.9s · 40 cr Try
- Recraft v428%Recraft · retired from voting12.6s · 45 cr Try
- Midjourney Niji 719%Midjourney · retired from voting108.6s · 160 cr Try
FAQ
- Which AI model renders this prompt best?
- GPT Image 2.5 Sunburst leads this prompt in the BudgetPixel AI Image Arena, winning 77% of its 13 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.