GPT Image Prompt
Ultra-realistic candid iPhone-style smartphone flash photograph, portrait orientation, 3:4 aspect ratio. Preserve the active LoRA identity reference as the absolute subject authority: the final woman must keep the exact same facial geometry, eye spacing, eye shape structure, nose, lips, jaw, cheeks, skull proportions, hair identity, skin tone, body geometry, shoulder-to-head relationship, shoulder width, torso-to-leg balance, waist thickness, hip width, hip-to-thigh transition, thigh volume, leg length ratio, body mass distribution, and recognizable physical presence from the LoRA reference. Do not slim, elongate, beautify, narrow, reshape, or replace the LoRA face or body with a generic attractive East Asian woman.
A young East Asian woman with the LoRA subject’s exact identity is walking quickly along a quiet modern urban street at night beside a long white tiled wall. She is caught in a spontaneous paparazzi-like smartphone moment: her body is still moving forward down the sidewalk while her head glances back over her shoulder toward the camera. She suddenly notices the phone and instinctively raises one open hand directly toward the lens to block the shot. The raised hand is large in the near foreground because it is closest to the phone, with natural palm tension, realistic finger spacing, visible knuckle structure, believable wrist articulation, and slight motion blur from the sudden movement. Her expression is alive and subtle, not stiff: direct eye contact through thin-framed glasses, a small playful uneven smile, relaxed mouth tension, and a caught-mid-reaction feeling. Do not copy any blank, dead-eyed, frozen, neutral, or mannequin expression from the LoRA reference.
She wears a sleek black sleeveless dress that follows the LoRA reference body truthfully without compressing the waist, narrowing hips, reducing thigh volume, lengthening legs, or creating a fashion-model silhouette. The dress has realistic fabric tension from walking motion and shoulder movement. She wears thin-framed glasses, small earrings, and carries a glossy black shoulder bag on one shoulder, with the strap resting physically on the body and the bag responding naturally to movement. Long dark hair consistent with the LoRA reference is lightly swept by wind, with a few strands blurred or displaced by motion and flash.
The setting is a dimly lit urban night street with a long white tiled wall beside her, dark asphalt pavement, distant streetlights, high-rise building hints, scattered soft city bokeh, and ordinary real-world clutter or imperfect background overlap. Keep the environment physically believable, not a luxury cinematic set: realistic wall scale, sidewalk scale, pavement texture, distant light falloff, grounded feet, natural contact shadows, and coherent perspective. The subject must feel physically inside the street scene, not composited onto it.
All expression, gesture, outfit, pose, camera angle, lighting, motion blur, environment, smartphone flash behavior, and realism details remain subordinate to the active LoRA identity reference. The final image must still be unmistakably the exact same LoRA subject in face and body, not a similar-looking substitute, not a beautified replacement, not a generic attractive version of the prompt, and not the same face placed onto a different body type.
Strong direct on-camera smartphone flash creates bright foreground illumination on the hand, face, glasses, upper body, and bag, with harsh falloff, clipped highlights in small areas, deep background shadows, uneven skin reflectance, realistic glasses reflections, and a raw late-night social-photo feel. Use authentic iPhone-style smartphone optics: handheld portrait framing, slight off-center composition, mild wide-angle perspective, imperfect horizon balance, casual crop behavior, small-sensor depth response, computational sharpening inconsistencies, localized noise, slight compression artifacts, and realistic low-light HDR limitations. The outstretched hand dominates part of the foreground and may partially obscure the face or body, but identity must remain recognizable. Motion blur should affect the raised hand, hair strands, and walking movement naturally, while preserving enough facial identity structure.
The photo should feel like an accidental candid smartphone flash snapshot from a late-night Tokyo-like street walk, not a staged fashion editorial. Preserve documentary realism, natural skin texture, realistic facial asymmetry, ordinary human posture, imperfect capture timing, environmental messiness, and believable contact physics. Avoid cinematic lighting perfection, luxury editorial polish, influencer beauty posing, seductive gaze, over-smoothed skin, artificial symmetry, perfect bokeh, DSLR lens compression, fake portrait-mode blur, over-clean HDR, porcelain skin, CGI textures, mannequin hands, fused fingers, duplicated fingers, floating feet, disconnected shadows, impossible wrist angles, unrealistic leg elongation, narrowed waist, narrowed hips, reduced thighs, generic model body, hyper-polished fashion photography, and any smartphone angle that changes the LoRA reference body identity.
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”.