Original document
GPT Image 2 · en
Scene:
A single vertical 9:21 ultra-long scroll painting inspired by detailed urban reportage, viewed as one continuous hand-painted lane from top to bottom. The composition flows through an old Beijing hutong from dawn at the top, midday in the middle, and evening lantern light near the bottom.
Subject:
A modern version of a historical city scroll: courtyard gates, gray brick walls, tiled roofs, bicycle repair stall, breakfast steam, chess table, schoolchildren, delivery carts, small groceries, residents leaning from balconies, and cats only as tiny background silhouettes if present. The lane must feel continuous and navigable, not a collage of unrelated vignettes.
Important details:
Extreme vertical 9:21 framing, dense but legible local scenes, clear spatial continuity, soft ink-and-gouache texture, accurate north China architectural details, atmospheric perspective, repeated alley geometry that guides the eye downward, hundreds of small believable objects without turning into noise.
Use case:
Independent Prompt Atlas image for testing extreme aspect ratio stability, high-density narrative planning, and world-knowledge scene detail.
Constraints:
No readable text, no watermark, no logos, no floating labels, no modern app UI, no frame, no broken perspective, no random fantasy elements, preserve exact 9:21 vertical aspect ratio.How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.
