Typography
How 20 AI image models render: “At dusk, a narrow flower shop with rain-dark brick, overflowing window boxes, and warm interior light; the invented name "EMBER VAULT" is rendered in red neon tubing above the entrance, reflected in puddles.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: is the text exactly right, spelled correctly, well set? 294 community votes so far.
Fresh prompt from the daily stream · voting closed 2026-09-20
Expected text: EMBER VAULT
20 text-to-image models rendered this typography prompt with the identical text at their default settings and a fixed seed. Recraft v4 is the community favourite so far, winning 74% of its 34 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 5.1s.
What to look for: correct spelling, placement and legibility, and type that looks printed or carved rather than pasted on.
- Recraft v474%Recraft · retired from voting12.8s · 45 cr Try
- GPT Image 2.5 Sunburst67%OpenAI30.4s · 28 cr Try
- Seedream 5.0 Pro65%ByteDance26.4s · 49 cr Try
- GPT Image 2.5 Flare61%OpenAI28.8s · 28 cr Try
- Krea 2 Large56%Krea38.3s · 80 cr Try
- Muse Image 1.052%Meta23.6s · 12 cr Try
- Recraft v4 Pro52%Recraft · retired from voting22.6s · 300 cr Try
- 46%16.7s · 50 cr Try
- ImagineArt 246%ImagineArt · retired from voting38.1s · 60 cr Try
- Nano Banana 2 Lite45%Google22.1s · 36 cr Try
- Ideogram v445%Ideogram15.3s · 40 cr Try
- ImagineArt 1.5 Pro45%ImagineArt · retired from voting54.1s · 55 cr Try
- Qwen Image 3.045%Alibaba15.8s · 32 cr Try
- GPT Image 241%OpenAI · retired from voting57.0s · 60 cr Try
- 5.1s · 80 cr Try
- Qwen Image 3.0 Pro39%Alibaba12.7s · 36 cr Try
- Kling v3 Omni36%Kuaishou27.2s · 35 cr Try
- Krea 2 Medium27%Krea31.5s · 40 cr Try
- Midjourney Niji 715%Midjourney · retired from voting103.0s · 160 cr Try
FAQ
- Which AI model renders this prompt best?
- Recraft v4 leads this prompt in the BudgetPixel AI Image Arena, winning 74% 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.