Typography
How 20 AI image models render: “A café chalkboard listing exactly three items: "Flat white 4", "Cardamom bun 5", "Soup of the day 9", in neat chalk handwriting.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: is the text exactly right, spelled correctly, well set? 274 community votes so far.
Core suite prompt · voting closed 2026-09-18
Expected text: Flat white 4 | Cardamom bun 5 | Soup of the day 9
20 text-to-image models rendered this typography prompt with the identical text at their default settings and a fixed seed. Recraft v4 Pro is the community favourite so far, winning 79% of its 29 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 4.8s.
What to look for: correct spelling, placement and legibility, and type that looks printed or carved rather than pasted on.
- Recraft v4 Pro79%Recraft · retired from voting13.7s · 300 cr Try
- ImagineArt 275%ImagineArt · retired from voting28.2s · 60 cr Try
- Seedream 5.0 Pro72%ByteDance69.1s · 49 cr Try
- Nano Banana 2 Lite68%Google21.8s · 36 cr Try
- GPT Image 2.5 Flare68%OpenAI23.8s · 40 cr Try
- ImagineArt 1.5 Pro67%ImagineArt · retired from voting55.5s · 55 cr Try
- Muse Image 1.061%Meta10.3s · 12 cr Try
- GPT Image 253%OpenAI · retired from voting45.0s · 60 cr Try
- 4.8s · 80 cr Try
- Ideogram v450%Ideogram13.6s · 40 cr Try
- Kling v3 Omni45%Kuaishou36.0s · 35 cr Try
- GPT Image 2.5 Sunburst40%OpenAI25.6s · 80 cr Try
- Krea 2 Medium38%Krea23.7s · 40 cr Try
- Qwen Image 3.0 Pro35%Alibaba11.8s · 36 cr Try
- Recraft v432%Recraft · retired from voting10.9s · 45 cr Try
- Qwen Image 3.032%Alibaba8.4s · 32 cr Try
- 29%12.2s · 50 cr Try
- Krea 2 Large28%Krea39.1s · 80 cr Try
- Midjourney Niji 711%Midjourney · retired from voting106.7s · 160 cr Try
FAQ
- Which AI model renders this prompt best?
- Recraft v4 Pro leads this prompt in the BudgetPixel AI Image Arena, winning 79% of its 29 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.