OpenZhenZhen AI Prompt Gallery · gpt-image-2.5 · image

A Century-Spanning Reunion on the Streets of New York

A Century-Spanning Reunion on the Streets of New York — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.

A Century-Spanning Reunion on the Streets of New York — AI image creation
A Century-Spanning Reunion on the Streets of New York

Prompt

Chinese prompt

身份依赖：执行时提供identity_reference，并使用固定参考ref-10a-v1统一两个时代中同一位明确成年女性的面部身份；本整理步骤不声称来源所说的上传参考图已附加或检查。创作一幅电影感照片写实人像：场景为纽约街道，画面左侧呈现带有帝国大厦的1930年代纽约，右侧过渡至现代纽约时代广场。画面中同一位成年女性在两个时代各出现一次，面部至上半身均清晰可见；年龄感、面部身份、发型身份和身体比例保持一致，只改变符合时代的服装与造型。左侧使用符合1930年代的纽约建筑、街面材质、车辆与时尚，并让帝国大厦出现在城市环境中。右侧表现当代时代广场，包括现代城市建筑、交通和发光数字广告屏。让两个环境在画面中心自然融合，不出现生硬分割线；建筑、路面、服装纹理与道具由旧到新逐渐过渡。两侧的她可以略微相望，营造穿越时空与自我重逢的含蓄感觉。采用电影级照片写实光线、自然而富有戏剧性的对比、丰富材质细节与空间氛围，同时保证两张面部都清楚。采用4:7竖幅构图。不承诺精确面部匹配或8K输出。不要额外人物、说明文字、标志或水印。

English prompt

Identity dependency: provide identity_reference at execution and use fixed reference ref-10a-v1 to guide the same clearly adult woman's face and identity in both eras; the source's conditional uploaded reference is not claimed attached or inspected during curation. Create one cinematic photorealistic portrait composition on a New York street, transitioning from 1930s New York with the Empire State Building on the left to modern Times Square on the right. Show the same adult woman once in each era, both fully readable from face through upper torso, with the same apparent age, facial identity, hair identity and body proportions; only era-appropriate wardrobe and styling change. On the left, use historically plausible 1930s New York architecture, street materials, vehicles and fashion, with the Empire State Building visible in the city context. On the right, depict contemporary Times Square with current urban architecture, modern transport and luminous digital billboards. Let the two environments merge organically through the center with no hard dividing line: architecture, paving, clothing textures and props transition gradually from old to new. Pose each version of the woman so she may glance toward the other side, creating a quiet sense of time travel and self-encounter. Use cinematic photorealistic lighting, dramatic but natural contrast, rich material detail and atmospheric depth while keeping both faces clear. Compose in 4:7 portrait. Do not promise an exact face match or 8K output. No extra people, captions, logos or watermark.

View API model

Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas