原文
GPT Image 2 · en
Scene:
Compose a square menu-design mockup that combines printed surfaces, structured page hierarchy, and appetizing food photography into one coherent brand-authored system. The page layouts should feel intentional and premium, with strong margins, modular blocks, and realistic print or board materials even though all textual zones remain unreadable. Use a clean square still life on a linen-covered table with one folded menu object, one plate edge, and soft service lighting.
Subject:
The menu should communicate category structure, hero dish placement, and brand tone through composition, image framing, typographic blocks, and page rhythm rather than through readable copy. Food imagery must stay believable and publication-ready. Treat the folded menu as a luxury hospitality object with unreadable but intentional text zones and subtle dish-image integration.
Important details:
This topic tests layout reasoning as much as food styling. Preserve clean grids, realistic paper behavior, credible shadows, and strong focal hierarchy between cover, spreads, dish photos, and placeholder text zones. Text-like marks can exist only as abstract or unreadable forms; clarity must come from design structure. This image should feel fully self-contained while still belonging to the same broader topic and commercial world. Paper thickness, fold precision, foil-like accent blocks, and spatial calm should make the menu feel expensive and ceremonial.
Use case:
Menu System single study 5/9, built for Prompt Atlas coverage of a food and beverage commercial visual system.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no readable text, no logos, no fake menu prices, no warped paper errors, no amateur flyer clutter, no chaotic collage, no obvious software UI chrome, no legible script, no wedding invitation look, no stationery clutter.使い方
- パラメータを含む元のプロンプト全体をコピーします。
- 被写体、文字、構図を自分の要件に置き換えます。
- 対応する外部ツールで使用し、利用条件と結果を確認します。
開始前の確認
- 参考画像の要否を確認してください。互換モデルの表示は、そのモデルでの生成実績を意味しません。
- 名前、数値、画像内の文字を確認します。モデルの版で結果は変わります。
