← All benchmarks
Fairy-Tale Poster
3 models compared · same input · same prompt
Run it on your idea
Run your own prompt across all 3 models
Don’t just compare ours — get every model’s take on your idea, side by side. Full-res, one click. No juggling 3 accounts.
Input & prompt
Input

Prompt template
Create a vertical 3:4 composition poster with the upper half preserving the original photograph {{subject:image}} with subtle, refined photographic color grading, extending the sky, ground, or environment as needed without distorting the subject. The lower half features a fairy-tale printmaking-style hand-drawn illustration of the most recognizable subject from the photograph, compressed into one or two clear, dominant masses with blunt, slightly awkward silhouettes, mildly exaggerated proportions, and essential edges, preserving identity and emotion. Derive a large-scale, low-value or high-contrast structural field from the original image for the background, creating structural storytelling through occlusion, cropping, engulfing forms, and partial reveals, using flat narrative perspective and picture-book page hierarchy. Single subject, no extra people.ChatGPT (GPT Image 2)
Try this →gpt-image-2
Nano Banana 2 (Gemini)
Try this →nano-banana-2
Flux Kontext Pro
Try this →flux-kontext-pro
Want to try this on YOUR photo?
Run the same prompt with any photo of your own — it's free.
Try the prompt →New benchmark every week
We put the latest models head to head on the same brief — no labels, you judge. Get each new benchmark in your inbox the morning it drops.