Photorealism
How 22 AI image models render: “A man is pushing his car, which is stuck in the mud, amid heavy rain in the countryside. The sky is gloomy, and some hail is also falling on the car and the ground. The big man has mud all over his trousers. Show him struggling.”
Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: could this be a photograph? Skin, light, lens behaviour. 161 community votes so far.
Community prompt by @sharierkhan · voting closed 2026-10-16
22 text-to-image models rendered this photorealism prompt with the identical text at their default settings and a fixed seed. Nano Banana 2.1 is the community favourite so far, winning 82% of its 11 head-to-head votes on this prompt. Grok Imagine Image Quality was the fastest at 4.7s.
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.
- Nano Banana 2.182%Google10.8s · 31 cr Try
- Midjourney V8.264%Midjourney122.6s · 160 cr Try
- GPT Image 2.5 Sunburst63%OpenAI41.6s · 40 cr Try
- Nano Banana 2 Lite63%Google16.4s · 36 cr Try
- Seedream 5.0 Pro60%ByteDance23.9s · 49 cr Try
- ImagineArt 260%ImagineArt · retired from voting34.0s · 60 cr Try
- Seedream 5.0 Flash55%ByteDance21.6s · 22 cr Try
- Ideogram v453%Ideogram14.3s · 40 cr Try
- Ideogram v4.550%Ideogram11.3s · 80 cr Try
- 4.7s · 80 cr Try
- Krea 2 Medium50%Krea19.8s · 40 cr Try
- FLUX 3 Image36%Black Forest Labs28.6s · 65 cr Try
- Krea 2 Large33%Krea30.4s · 80 cr Try
- Qwen Image 3.033%Alibaba7.7s · 32 cr Try
- 31%21.4s · 50 cr Try
- Qwen Image 3.0 Pro27%Alibaba10.2s · 36 cr Try
- Muse Image 1.023%Meta11.3s · 12 cr Try
- GPT Image 2.5 Flare14%OpenAI21.9s · 40 cr Try
- Kling v3 Omni10%Kuaishou61.3s · 35 cr Try
- GPT Image 2OpenAI · retired from voting47.9s · 60 cr Try
- FLUX 3 Image KleinBlack Forest Labs5.0s · 12 cr Try
FAQ
- Which AI model renders this prompt best?
- Nano Banana 2.1 leads this prompt in the BudgetPixel AI Image Arena, winning 82% 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.