
Original document
GPT Image 2 · en
Scene:
Create a square dependent campaign image in the exact same festive environment, preserving the same lacquer red, warm gold, floral freshness, and premium depth from the reference image, but make the composition cleaner and more retail-counter oriented. The frame should feel obviously related to the anchor while shifting toward a polished beauty or lifestyle merchandising surface suitable for campaign rollout, storefront crops, and social reuse.
Subject:
Feature premium beauty-like products, trays, gift boxes, blossom accents, and celebratory forms arranged into one coherent Lunar New Year display. Preserve exact palette, lighting direction, packaging geometry, and festive energy from the reference model while changing the assortment emphasis toward counter merchandising. The result should remain luxurious, brand-neutral, and useful for retail hero imagery or high-frequency seasonal tiles.
Important details:
Preserve exact campaign identity from the reference image. The celebratory energy should come from rich materials, spacing, and color rather than readable slogans. Use product silhouettes, reflective surfaces, and gifting logic to show category specificity. Keep the balance between abundance and elegance disciplined enough that the image reads premium rather than noisy or mass-market generic.
Use case:
LNY Rollout series study 2/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, preserve exact campaign identity from the reference image, no readable labels, no mascot dolls, no glitter explosion, no fashion model, no logo embossing, no fireworks, no theme-park saturation.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.





