Typography
How 22 AI image models render: “A brass luggage tag hangs from a weathered sea chest in a harbor loft, its face bearing “MIST HARBOR” in deeply stamped blue capitals; ropes and salt-crusted fittings surround it, with no other writing visible.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: is the text exactly right, spelled correctly, well set? 171 community votes so far.
Fresh prompt from the daily stream · voting closed 2026-10-18
Expected text: MIST HARBOR
22 text-to-image models rendered this typography prompt with the identical text at their default settings and a fixed seed. FLUX 3 Image Klein is the community favourite so far, winning 67% of its 9 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.
- Muse Image 1.067%Meta18.3s · 12 cr Try
- FLUX 3 Image Klein67%Black Forest Labs7.0s · 12 cr Try
- Midjourney V8.264%Midjourney94.3s · 160 cr Try
- 5.1s · 80 cr Try
- Ideogram v4.554%Ideogram13.6s · 80 cr Try
- Seedream 5.0 Flash50%ByteDance21.6s · 22 cr Try
- Seedream 5.0 Pro50%ByteDance25.8s · 49 cr Try
- GPT Image 2.5 Sunburst50%OpenAI76.9s · 40 cr Try
- Krea 2 Large50%Krea95.7s · 80 cr Try
- GPT Image 2.5 Flare42%OpenAI25.8s · 40 cr Try
- 42%13.8s · 50 cr Try
- Nano Banana 2.142%Google9.5s · 31 cr Try
- Nano Banana 2 Lite38%Google61.7s · 36 cr Try
- Qwen Image 3.036%Alibaba8.2s · 32 cr Try
- Qwen Image 3.0 Pro33%Alibaba10.9s · 36 cr Try
- FLUX 3 Image33%Black Forest Labs23.5s · 65 cr Try
- Kling v3 Omni29%Kuaishou61.8s · 35 cr Try
- Ideogram v425%Ideogram12.4s · 40 cr Try
- Krea 2 Medium20%Krea26.9s · 40 cr Try
- ImagineArt 210%ImagineArt · retired from voting98.4s · 60 cr Try
- GPT Image 2OpenAI · retired from voting139.7s · 60 cr Try
FAQ
- Which AI model renders this prompt best?
- FLUX 3 Image Klein leads this prompt in the BudgetPixel AI Image Arena, winning 67% of its 9 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.