How 22 AI image models render: “Exactly three green apples and two yellow pears on a white plate, the plate on a blue tablecloth, nothing else on the table.”
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 2026-09-16
Checklist: exactly three green apples · exactly two yellow pears · white plate · blue tablecloth · nothing else on the table
22 text-to-image models rendered this prompt adherence prompt with the identical text at their default settings and a fixed seed. ImagineArt 1.5 Pro is the community favourite so far, winning 100% of its 12 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 5.3s.
What to look for: the right count of objects, the right spatial relationships, the stated colours and attributes, and nothing added.
- ImagineArt 1.5 Pro100%ImagineArt · retired from voting50.0s · 55 cr Try
- Nano Banana 2 Lite85%Google19.9s · 36 cr Try
- Muse Image 1.072%Meta8.1s · 12 cr Try
- Recraft v471%Recraft · retired from voting13.9s · 45 cr Try
- Krea 2 Medium69%Krea27.1s · 40 cr Try
- Kling v3 Omni63%Kuaishou31.4s · 35 cr Try
- ImagineArt 257%ImagineArt · retired from voting31.5s · 60 cr Try
- GPT Image 2.5 Sunburst53%OpenAI30.5s · 80 cr Try
- Qwen Image 3.0 Pro47%Alibaba9.8s · 36 cr Try
- Recraft v4 Pro43%Recraft · retired from voting20.8s · 300 cr Try
- Qwen Image 3.042%Alibaba10.5s · 32 cr Try
- Ideogram v436%Ideogram16.3s · 40 cr Try
- 5.3s · 80 cr Try
- GPT Image 2.5 Flare33%OpenAI30.9s · 40 cr Try
- 31%8.9s · 50 cr Try
- Krea 2 Large22%Krea72.1s · 80 cr Try
- Seedream 5.0 Pro18%ByteDance30.8s · 49 cr Try
- GPT Image 214%OpenAI · retired from voting57.7s · 60 cr Try
- Midjourney Niji 713%Midjourney · retired from voting115.3s · 160 cr Try
- Midjourney v7Midjourney · retired from voting145.3s · 160 cr Try
FAQ
- Which AI model renders this prompt best?
- ImagineArt 1.5 Pro leads this prompt in the BudgetPixel AI Image Arena, winning 100% of its 12 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.