GPT Image Prompt
A realistic vertical iPhone-style nighttime street photo using the active LoRA identity reference as the absolute identity authority for the woman’s face and full body. Preserve the exact same LoRA subject identity: recognizable facial geometry, eye spacing and shape, nose/lip/jaw/cheek structure, skull proportions, light skin tone, hair identity, long dark-rooted blonde-gray hair tied high with loose wavy lengths around her shoulders, same narrow shoulder width, same shoulder-to-head relationship, same torso length, same torso-to-leg balance, same waist thickness, same hip width, same thigh volume, same leg length ratio, same body mass distribution, and the same grounded adult physical presence. Do not make her slimmer, taller, curvier, longer-legged, narrower-waisted, narrower-hipped, reduced-thigh, more model-like, or more influencer-like than the active LoRA reference.
She is standing on a quiet city street at night in a slight three-quarter body turn, weight shifted naturally through one hip, with believable balance and grounded feet on the road surface. Her right hand is raised near her temple as if casually adjusting her hair, with realistic wrist articulation, natural finger spacing, believable hand tension, and no fused or duplicated fingers. Her left arm hangs lower, holding a small bright pink handbag near her thigh with realistic grip pressure, finger contact, object weight, and natural arm relaxation. Her posture feels caught between posing and moving, not a fashion-pose, with slight human asymmetry and casual body tension.
Her expression is subtle, alive, and real rather than frozen: a brief camera-aware glance, relaxed eyelids, softly parted pursed lips, and a mildly reserved look caught mid-moment. Preserve the source-prompt human moment while keeping the LoRA facial structure unmistakably the same person. Do not copy blank LoRA-reference stiffness, dead eyes, mannequin neutrality, or a commercial model smile.
She wears a black cropped short-sleeve T-shirt with small yellow chest text, dark high-waisted distressed denim shorts with side cutout ring-chain details and frayed edges, visible thigh tattoos, and partial torso and arm tattoos. The clothing must sit naturally on the LoRA reference body geometry, with realistic fabric tension, gravity, folds, waist contact, and no reshaping of her torso, waist, hips, thighs, or leg ratio. Tattoos should appear naturally integrated into skin texture and smartphone reconstruction, not overly sharp or artificially printed.
The scene is an ordinary handheld phone capture on a quiet urban road at night, with lane markings, a curb edge, dark trees, storefront lights, and distant signage behind her. Preserve believable street scale, real curb and road geometry, physically plausible human-to-environment proportions, and attached contact shadows under her feet. The environment should feel casually real, slightly cluttered, imperfect, and not staged: uneven background lights, practical street darkness, ordinary storefront glow, imperfect object spacing, and natural visual contamination.
All expression, gesture, outfit, pose, camera behavior, lighting, street setting, smartphone realism, and visual imperfections remain subordinate to the active LoRA identity reference. The final subject must still be unmistakably the exact same LoRA woman in both face and body, not a similar-looking substitute, not a beautified replacement, not a generic attractive version, and not the same face placed onto a different body type.
Shoot as a realistic vertical iPhone-style nighttime street photo with casual imperfect framing like a social phone photo. Use mild wide-angle smartphone perspective without body beautification: no leg elongation, no waist slimming, no hip narrowing, no thigh reduction, no flattering distortion that changes identity. The subject is brightened by low-light smartphone auto-exposure while the street remains dark, with compressed shadows, slight warm highlight bloom from background lights, practical exposure imbalance, mild HDR reconstruction, localized sharpness variation, soft hair edges, subtle low-light noise, denoise smearing, compression artifacts, imperfect edge reconstruction, and uneven smartphone detail recovery.
Avoid cinematic lighting, DSLR depth blur, luxury editorial composition, perfect symmetry, fake HDR polish, over-smoothed skin, plastic skin, over-sharpened tattoos, mannequin stiffness, blank stare, influencer-style beautification, fashion-model body reshaping, artificial waist compression, leg elongation, reduced thigh volume, narrowed hips, fake social-media perfection, sterile background cleanliness, floating feet, disconnected shadows, impossible hand anatomy, fake handbag grip, and any camera angle or pose interpretation that changes the active LoRA face/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”.