How 21 AI image models render: “Create a cinematic, photorealistic literary-fiction still at a modest gas station in late-afternoon light. Capture the instant a middle-aged man has started walking around a gas pump toward a tall, hyper-muscular and athletic woman filling her car with gas. Keep the camera close to the man's subjective perspective: part of his shoulder or arm may frame the foreground, while the pump divides the composition and suggests the choice he is making. His posture should show him slightly tense, hesitant, not threatening. The woman is around six feet tall, with a strong, confident presence, wearing a soft, close-fitting T-shirt, casual shorts, and ordinary sandals; visible only as part of the scene rather than as a close-up or focal image. Give her agency: she has noticed John approaching and is looking toward him with a calm, assessing expression, neither inviting nor frightened. Include grounded gas-station details—fuel pump, parked car, sun-faded signage, and the open road.”

Same prompt, sent verbatim, at each model's default settings and a fixed seed. Judge: could this be a photograph? Skin, light, lens behaviour. 228 community votes so far.

Community prompt by @underman · voting closed 2026-11-14

21 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 72% of its 18 head-to-head votes on this prompt. FLUX 3 Image Klein was the fastest at 6.2s.

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.

FAQ

Which AI model renders this prompt best?

GPT Image 2.5 Flare leads this prompt in the BudgetPixel AI Image Arena, winning 72% of its 18 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.