
Original document
GPT Image 2 · en
Scene:
Create a square premium florist-counter image with soft afternoon window light, blush and wine-red florals, and polished stone or lacquer work surfaces. The scene should feel like a campaign image for a high-end flower studio rather than a market stall. Keep the composition controlled, layered, and commercially clear, with enough open space and object separation to support hero use in social, gifting guides, or seasonal landing pages.
Subject:
Show bouquets in different stages of preparation, satin ribbon, tissue, a wrapped stem bundle, clean scissors, and one elegantly boxed gift integrated into the florist environment. The frame should communicate romantic gifting through floral craft and premium service. Focus on bouquet structure, wrapping quality, and counter styling rather than on people, readable notes, or wedding symbolism.
Important details:
Use tasteful romance cues only: nuanced petals, calm ribbon folds, realistic wrapping paper, and a restrained luxurious palette. The image should feel giftable and contemporary, not cliché. Keep stems, petals, paper folds, and counter reflections materially persuasive. Everything should read as a usable brand visual for Valentines promotion without requiring text overlays or overt seasonal gimmicks.
Use case:
Valentines single study 5/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no readable cards, no visible customer faces, no wedding veil cues, no giant heart balloons, no flower-shop logo, no cluttered stems on the floor, no cartoon décor, no oversaturated magenta.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.





