How 22 AI image models render: “Realistic. Three scenes in one. 1) A room filled with many types of candy. 2) A room filled with many kinds of animals. 3) A room filled with all kinds of money. Above each room, a list names what is inside.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: every stated element present, correct count, correct placement. 136 community votes so far.
Community prompt by @blanknamebp · voting closed 2026-10-17
Checklist: 3 scenes in one image · 1 room filled with many types of candy · 1 room filled with many kinds of animals · 1 room filled with all kinds of money · A list above each room names what is inside
22 text-to-image models rendered this prompt adherence prompt with the identical text at their default settings and a fixed seed. Midjourney V8.2 is the community favourite so far, winning 90% of its 10 head-to-head votes on this prompt. FLUX 3 Image Klein was the fastest at 4.9s.
What to look for: the right count of objects, the right spatial relationships, the stated colours and attributes, and nothing added.
- Midjourney V8.290%Midjourney91.0s · 160 cr Try
- GPT Image 2.5 Sunburst75%OpenAI50.3s · 40 cr Try
- Seedream 5.0 Pro69%ByteDance69.1s · 49 cr Try
- FLUX 3 Image Klein60%Black Forest Labs4.9s · 12 cr Try
- Nano Banana 2.153%Google16.4s · 31 cr Try
- GPT Image 2.5 Flare50%OpenAI40.5s · 40 cr Try
- Nano Banana 2 Lite50%Google59.2s · 36 cr Try
- ImagineArt 250%ImagineArt · retired from voting29.6s · 60 cr Try
- Ideogram v4.545%Ideogram17.0s · 80 cr Try
- Ideogram v443%Ideogram16.5s · 40 cr Try
- 38%23.7s · 50 cr Try
- Seedream 5.0 Flash38%ByteDance26.4s · 22 cr Try
- Muse Image 1.038%Meta13.8s · 12 cr Try
- FLUX 3 Image33%Black Forest Labs25.6s · 65 cr Try
- 6.9s · 80 cr Try
- Kling v3 Omni33%Kuaishou62.7s · 35 cr Try
- Qwen Image 3.021%Alibaba7.1s · 32 cr Try
- Qwen Image 3.0 Pro21%Alibaba10.8s · 36 cr Try
- Krea 2 Medium10%Krea21.7s · 40 cr Try
- Krea 2 Large9%Krea38.2s · 80 cr Try
- GPT Image 2OpenAI · retired from voting67.6s · 60 cr Try
FAQ
- Which AI model renders this prompt best?
- Midjourney V8.2 leads this prompt in the BudgetPixel AI Image Arena, winning 90% of its 10 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.