Architecture & interiors
How 22 AI image models render: “Exterior of a contemporary alpine research lodge in restrained regional modernism, rough granite walls, larch screens, steep standing-seam roof, crisp snow-bright morning light, viewed from below across a mountain footpath, sound structural geometry.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: straight lines, materials, believable space. 163 community votes so far.
Fresh prompt from the daily stream · voting closed 2026-10-03
22 text-to-image models rendered this architecture & interiors prompt with the identical text at their default settings and a fixed seed. GPT Image 2 is the community favourite so far, winning 75% of its 16 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 5.2s.
What to look for: straight verticals, plausible structure, consistent perspective and correct materials.
- GPT Image 275%OpenAI275.6s · 60 cr Try
- GPT Image 2.5 Flare72%OpenAI129.4s · 40 cr Try
- ImagineArt 1.5 Pro67%ImagineArt52.2s · 55 cr Try
- 65%15.9s · 50 cr Try
- Nano Banana 2 Lite64%Google138.6s · 36 cr Try
- Seedream 5.0 Pro61%ByteDance74.7s · 49 cr Try
- Seedream 5.0 Flash50%ByteDance32.1s · 22 cr Try
- Muse Image 1.048%Meta21.0s · 12 cr Try
- Kling v3 Omni46%Kuaishou27.7s · 35 cr Try
- Ideogram v444%Ideogram14.6s · 40 cr Try
- Qwen Image 3.0 Pro39%Alibaba10.4s · 36 cr Try
- Recraft v438%Recraft · retired from voting11.9s · 45 cr Try
- GPT Image 2.5 Sunburst38%OpenAI414.9s · 40 cr Try
- ImagineArt 238%ImagineArt121.4s · 60 cr Try
- Recraft v4 Pro38%Recraft12.1s · 300 cr Try
- Qwen Image 3.029%Alibaba7.1s · 32 cr Try
- Ideogram v4.529%Ideogram12.8s · 80 cr Try
- 5.2s · 80 cr Try
- Krea 2 Large24%Krea28.5s · 80 cr Try
- Krea 2 Medium22%Krea15.9s · 40 cr Try
- Midjourney Niji 717%Midjourney · retired from voting268.7s · 160 cr Try
FAQ
- Which AI model renders this prompt best?
- GPT Image 2 leads this prompt in the BudgetPixel AI Image Arena, winning 75% of its 16 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.