

Create an ultra-realistic creative portrait of a young woman standing indoors against a clean white wall, seamlessly interacting with a large hand-drawn black-and-white pencil sketch of her standing beside her. The real woman stands on the right side, wearing an oversized black distressed sweatshirt, loose dark cargo-style jeans with stitched panel details, and black-and-white canvas sneakers. She has long straight pastel pink hair and a playful, slightly surprised smile, looking directly at the camera. Her arms are raised outward with both palms facing upward in a cute “what’s going on?” pose. Behind and slightly to the left of her, create a large expressive hand-drawn pencil sketch version of the same woman. The sketch should closely match her pose, proportions, hairstyle, facial expression, clothing silhouette, and hand positions, drawn with loose rough graphite strokes, visible cross-hatching, imperfect lines, and an authentic notebook-sketch aesthetic. Make the sketch appear physically drawn directly onto the wall, with parts of the sketch slightly overlapping the real woman to create a surreal 2D-to-3D interaction. Add playful hand-drawn doodles around the sketch, including stars, hearts, motion lines, small circles, and expressive marks. Minimal white studio background, soft natural lighting, subtle floor shadow, realistic skin and fabric texture, detailed hair strands, authentic photography, strong contrast between the monochrome pencil drawing and the colorful real person, whimsical editorial photography, clever visual storytelling, seamless integration, highly detailed, photorealistic, vertical composition, 4:5 aspect ratio.
This image was generated with GPT Image 2.5 by OpenAI. The prompt on this page is the exact text used to create it.
The example was generated with GPT Image 2.5 using: 561:701 · Digital Art · Realistic · Portrait. 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.