
原始文档
GPT Image 2 · en
Scene:
Compose a square feast or banquet campaign image with abundant but controlled table choreography, rich servingware, and a celebratory atmosphere that feels suitable for hospitality campaigns, seasonal dining promotions, or event editorials. The frame should manage many elements without losing hierarchy, elegance, or appetite appeal. Shoot from overhead on a large table section with dessert plates, fruit platters, and serving tools arranged into a balanced square composition.
Subject:
The food must look substantial and celebratory, with believable multi-dish coordination, serving logic, and table rituals that suggest a real banquet rather than a random buffet pile. Use repeated forms and color grouping to make the spread feel abundant yet controlled and campaign-friendly.
Important details:
This topic depends on composition reasoning. Preserve a strong centerpiece, readable secondary dishes, coherent servingware, and lighting that supports depth across the table. Celebration should emerge from abundance, color, and ritual objects, not from gimmicks or branded holiday symbols. Everything must feel premium, coordinated, and logistically plausible. This image should feel fully self-contained while still belonging to the same broader topic and commercial world. Spacing, platter size, and color rhythm are critical. The scene should feel celebratory, premium, and clear enough to support hospitality promotion.
Use case:
Festive Feast single study 8/9, 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 generic wedding décor, no buffet-chaos clutter, no novelty party props, no copied holiday mascots, no visible faces dominating the frame, no candy-shop chaos, no children’s party props, no readable dessert labels.使用方法
- 复制完整原始提示词,并保留其中的参数。
- 按需求替换主体、文字和构图要求。
- 在兼容的外部工具中使用,核对当前使用条件并检查结果。
开始前需了解
- 先确认是否需要参考图。标注兼容模型,不代表该示例由这个模型生成。
- 核对名称、数字与画面文字;不同模型版本的结果可能不同。





