
Create an ultra-realistic, cinematic night portrait of a beautiful young woman sitting casually on a rooftop ledge overlooking a breathtaking city skyline at night. A gigantic glowing golden full moon dominates the sky directly behind her, surrounded by dramatic dark clouds and a subtle golden celestial glow. She has long, naturally flowing dark-brown wavy hair, delicate facial features, and a calm, dreamy expression while looking slightly to the side. She wears a fitted white crop top, an oversized dark blue denim jacket draped loosely over her shoulders, relaxed beige wide-leg trousers, and clean white sneakers. Her pose is relaxed and effortlessly stylish, sitting on the concrete rooftop edge with one knee raised and one leg hanging naturally over the ledge. One hand rests beside her while the other rests casually on her raised knee. Background: Stunning nighttime city skyline with illuminated skyscrapers, a glowing suspension bridge, sparkling golden city lights, and reflections shimmering across the river. Lighting & Atmosphere: Magical golden moonlight illuminating her hair and face, cinematic shadows, warm ambient city lighting, dramatic clouds, dreamy midnight atmosphere, luxury editorial photography. Camera & Quality: Full-body composition, vertical 9:16 portrait, shot on Sony A7R V, 85mm lens, f/1.8, shallow depth of field, realistic skin texture, highly detailed clothing, professional color grading, HDR, 8K ultra-realistic photography. Important: Preserve the exact facial identity if using a reference image. Natural body proportions, realistic hands and fingers, anatomically accurate pose, no extra limbs, no text, no watermark.
This image was generated with GPT Image 2.6 by OpenAI. The prompt on this page is the exact text used to create it.
The example was generated with GPT Image 2.6 using: 941:1672 · Cinematic · Portrait · Cityscape. Reusing these settings gives the closest match.
Copy the prompt as-is first, then replace only the subject while keeping the style, lighting, and composition keywords. Generate a few variations — outputs differ between runs.
Yes. Prompts work best as starting points — adjust descriptors, aspect ratio, or style tags to make the output your own.