Photorealism
How 23 AI image models render: “A husband and wife are shopping in a large, well-lit supermarket. They are surprised by what is for sale in the Christmas aisle.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: could this be a photograph? Skin, light, lens behaviour. 97 community votes so far.
Community prompt by @philbland · voting closed 2026-10-05
23 text-to-image models rendered this photorealism prompt with the identical text at their default settings and a fixed seed. GPT Image 2.5 Flare is the community favourite so far, winning 91% of its 11 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 6.6s.
What to look for: believable skin and fabric, physically plausible light and shadow, lens behaviour such as depth of field and motion blur, and no waxy or over-smoothed surfaces.
- GPT Image 2.5 Flare91%OpenAI37.6s · 40 cr Try
- GPT Image 2.5 Sunburst64%OpenAI48.5s · 40 cr Try
- Nano Banana 2 Lite64%Google21.6s · 36 cr Try
- Seedream 5.0 Pro57%ByteDance25.4s · 49 cr Try
- Qwen Image 3.0 Pro50%Alibaba12.5s · 36 cr Try
- 45%22.6s · 50 cr Try
- Krea 2 Medium45%Krea20.4s · 40 cr Try
- ImagineArt 1.5 Pro44%ImagineArt52.7s · 55 cr Try
- Recraft v4 Pro44%Recraft · retired from voting15.5s · 300 cr Try
- 6.6s · 80 cr Try
- Kling v3 Omni38%Kuaishou57.2s · 35 cr Try
- Muse Image 1.033%Meta13.7s · 12 cr Try
- GPT Image 230%OpenAI53.8s · 60 cr Try
- ImagineArt 229%ImagineArt31.0s · 60 cr Try
- Recraft v429%Recraft · retired from voting15.5s · 45 cr Try
- Ideogram v422%Ideogram17.1s · 40 cr Try
- Qwen Image 3.022%Alibaba6.9s · 32 cr Try
- Seedream 5.0 Flash20%ByteDance33.1s · 22 cr Try
- Krea 2 Large18%Krea27.3s · 80 cr Try
- Ideogram v4.5Ideogram13.6s · 80 cr Try
- Midjourney Niji 7Midjourney · retired from voting108.5s · 160 cr Try
- FLUX 3 ImageBlack Forest Labs23.9s · 32 cr Try
FAQ
- Which AI model renders this prompt best?
- GPT Image 2.5 Flare leads this prompt in the BudgetPixel AI Image Arena, winning 91% of its 11 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.