Prompt adherence
How 20 AI image models render: “A realistic multi-panel composition with 4–7 panels, each depicting an adult enjoying a wonderful day in a distinct everyday setting.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: every stated element present, correct count, correct placement. 238 community votes so far.
Community prompt by @blanknamebp · voting closed 2026-09-24
Checklist: 4–7 panels · each panel shows an adult · each adult is having a good day
20 text-to-image models rendered this prompt adherence 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 85% of its 34 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 6.6s.
What to look for: the right count of objects, the right spatial relationships, the stated colours and attributes, and nothing added.
- GPT Image 2.5 Sunburst85%OpenAI46.3s · 28 cr Try
- ImagineArt 1.5 Pro83%ImagineArt · retired from voting55.9s · 55 cr Try
- Nano Banana 2 Lite70%Google15.3s · 36 cr Try
- GPT Image 2.5 Flare63%OpenAI38.6s · 28 cr Try
- Recraft v460%Recraft · retired from voting16.9s · 45 cr Try
- ImagineArt 257%ImagineArt · retired from voting29.0s · 60 cr Try
- Muse Image 1.057%Meta19.6s · 12 cr Try
- Seedream 5.0 Pro50%ByteDance68.7s · 49 cr Try
- 6.6s · 80 cr Try
- Kling v3 Omni42%Kuaishou31.5s · 35 cr Try
- 39%23.9s · 50 cr Try
- Qwen Image 3.0 Pro37%Alibaba11.2s · 36 cr Try
- Ideogram v433%Ideogram14.4s · 40 cr Try
- Krea 2 Medium26%Krea14.9s · 40 cr Try
- Qwen Image 3.025%Alibaba7.5s · 32 cr Try
- Midjourney Niji 720%Midjourney · retired from voting122.3s · 160 cr Try
- Krea 2 Large10%Krea32.8s · 80 cr Try
- Recraft v4 ProRecraft · retired from voting23.3s · 300 cr Try
- GPT Image 2OpenAI · retired from voting67.4s · 60 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 85% of its 34 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.