
原始文档
GPT Image 2 · en
Scene:
A vertical 9:16 soft album portrait sheet with the same exact 3 by 3 grid geometry as the reference: nine equal cells, thin white gutters, clean edges, no labels, and a quiet editorial rhythm. The background becomes pale cream and powder blue fabric, with diffused window light, gentle film grain, and a calm photobook mood. Every cell should feel like part of one album package rather than nine unrelated snapshots.
Subject:
Use the reference image as the primary guide for the same young performer's face, chin-length black hair, soft fringe, dark almond eyes, beauty mark below the left eye, lip shape, and facial proportions. The styling changes to a gentle album concept: white cotton shirt, light blue cardigan, cream scarf, pearl-like stud earring, pale denim, soft knit vest, and simple natural makeup. The nine expressions are sleepy calm, shy smile, wide laughter, quiet sadness, curious raised eyebrow, warm direct gaze, cheek-in-hand thoughtfulness, surprised open eyes, and serene profile.
Important details:
Identity consistency is the main challenge. The viewer must believe every panel shows the same person even as expression, head angle, crop, hand placement, and outfit layers shift. Keep the grid mathematically clean, with all gutters straight and all cells equally sized. Maintain consistent skin tone, eye shape, hair length, and beauty mark placement. Avoid magazine text, sticker decorations, duplicate faces outside the grid, or heavy retouching that erases the subject's stable features.
Use case:
Nine-Expression Portrait Grid series study 3/8, testing reference-based identity consistency in a soft album concept.
Constraints:
No readable text, no watermark, no logos, preserve exact 9:16 vertical aspect ratio, no brand marks, no changed face identity, no extra subjects, no broken cell sizes, no mismatched beauty mark, no distorted hands.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。





