GPT Image Prompt
Use the active LoRA identity reference as the absolute authority for the final subject’s face and full-body identity. Preserve the exact same young woman, including her recognizable facial geometry, soft side-profile proportions, fair skin tone, long dark straight hair, shoulder width, shoulder-to-head relationship, torso length, torso-to-leg balance, waist thickness, hip width, thigh volume, leg length ratio, body mass distribution, and the same natural physical presence. Do not beautify, slim, elongate, narrow, or redesign her body or face. Body geometry is identity, not styling.
Create a realistic handheld iPhone-style night photo of this same woman at an outdoor waterfront railing. She is standing with her body turned mostly away from the camera toward the water, while her head turns left back over her shoulder, chin slightly lifted. Her expression is a quiet real-life over-shoulder moment: eyes glancing slightly upward and off-camera to the left, lips softly closed with faint natural tension, calm and slightly distracted, alive and natural rather than posed, blank, or mannequin-like. Both hands remain physically anchored on the black railing at opposite sides, arms extended back with relaxed elbows, posture upright with a slight backward lean and a natural hip weight shift. Keep the hand anatomy believable, with correct finger structure, realistic grip placement, and natural pressure/contact on the railing. She wears the same light pink cropped ruched short-sleeve top and the same light cream fitted mini skirt with a soft wavy hem, plus a thin necklace and a dark hair tie or bracelet on her right wrist. Preserve natural low-light fabric softness, small wrinkles, and believable clothing tension rather than editorial polish.
The setting is a casual urban waterfront or marina balcony at night, with dark water behind her, distant city lights, a skyline, and soft reflections on the water. The scene should feel socially ordinary and physically real, with plausible railing geometry, believable subject-to-environment scale, grounded posture, and realistic spatial depth.
All expression, gesture, outfit, pose, scene, camera, lighting, and realism behavior remain subordinate to the active LoRA identity reference. The final image must still be unmistakably the exact same LoRA subject in both face and body, not a similar-looking substitute, not a beautified replacement, and not a generic attractive version of the prompt. Preserve the source-prompt human moment while keeping the LoRA facial structure and body identity fully intact.
Shoot it like a spontaneous social snapshot from behind at close standing distance with imperfect vertical smartphone framing, slight handheld imbalance, mild iPhone wide-angle perspective, and ordinary human-operated camera behavior. Keep the composition casual rather than optimized. Simulate believable low-light smartphone behavior: uneven sharpness, compressed shadows, slight highlight bloom, subtle noise in the sky and water, mild texture instability, slight computational smoothing on skin and hair, and realistic low-light detail inconsistency. Let the background remain slightly unstable and imperfect rather than crisp or cinematic.
Favor ordinary realism over beauty. Avoid influencer aesthetics, luxury-nightlife polish, glossy fashion energy, cinematic lighting, over-clean HDR, artificial subject isolation, excessive bokeh, fake depth masking, perfect symmetry, over-posed body language, mannequin expression, floating hands, disconnected railing contact, unrealistic long legs, narrowed waist, narrowed hips, reduced thigh volume, or any smartphone angle that flatters the body away from the LoRA reference. The image should feel like a real casual iPhone night capture, not a staged or idealized AI photo.
How to use: Copy and paste this prompt into ChatGPT, Gemini, Nano Banana, or your favorite AI image tool. If it mentions LoRA, reference image, uploaded photo, or identity reference, upload your own photo first. No reference photo? Replace the character with a clear age, gender, and nationality, such as “25-year-old Japanese woman” or “30-year-old American man”.