GPT Image Prompt
Photorealistic casual iPhone-style vertical street photo of the exact same real woman from the active LoRA identity reference, with strict preservation of her full face and body identity before anything else: same recognizable facial geometry, eye spacing, eye shape structure, nose–lip–jaw relationship, oval face shape, cheek and chin structure, light natural skin tone, long dark straight hair identity, shoulder-to-head relationship, shoulder width, torso length, torso-to-leg balance, waist thickness, hip width, thigh volume, leg length ratio, body mass distribution, grounded adult height impression, and relaxed physical presence. Do not beautify, slim, elongate, reshape, or replace her body with a generic attractive version. The final subject must remain the exact LoRA woman, not a similar-looking substitute.
She is standing close to the camera on a muted ordinary urban street, torso leaning diagonally to one side with natural body weight and believable balance, shoulders uneven, head tilted strongly, one arm lowered near a foreground railing without forced posing. Her hair is blown loosely by wind across part of her face and over her shoulders, with imperfect strands and casual movement. Her expression stays subtle and alive in the wind-shifted moment: relaxed heavy eyelids, a brief camera-aware glance, softly parted lips, and a calm distracted look that matches the tilted head and leaning body. Preserve the source-prompt human moment; do not copy any blank LoRA-reference stiffness, dead-eyed stare, frozen mouth, mannequin neutrality, seductive influencer gaze, or commercial smile.
Preserve the outfit exactly in a realistic worn way: off-white cropped long-sleeve top with a wide neckline slipped off the shoulders, elastic lower band, lightweight slightly sheer sleeves, natural fabric wrinkles and tension from the leaning posture, low-rise faded light blue jeans sitting loosely at the hips, visible thin white string detail, and a silver chain shoulder strap running along one side. Clothing must follow gravity, body contact, and natural fabric compression without reshaping her reference body silhouette.
The background is an ordinary shaded city street with a parked vehicle, dark building façades, graffiti, road surface, railing, and practical urban clutter. Keep the environment physically believable, slightly messy, asymmetrical, and casually captured, with realistic human-to-environment scale, grounding, shadows, object contact, and spatial continuity. The foreground railing should feel physically present near her lowered arm, not floating or pasted in. Feet, lower body, clothing, shadows, and street surface must remain grounded and weight-aware if visible.
All expression, gesture, outfit, pose, camera behavior, lighting, realism treatment, and environmental details remain subordinate to the active LoRA identity reference. The final image must still be unmistakably the exact same LoRA subject in both face and full body identity, not a beautified replacement, not a fashion-model reinterpretation, not a slimmer or longer-legged version, and not a generic attractive woman matching the prompt.
Vertical handheld iPhone-style framing at close social-photo distance, slightly imperfect crop, mild wide-angle phone perspective, casual off-center composition, subtle handheld tilt, realistic smartphone edge distortion only where it does not alter her body identity. Subdued overcast daylight, shaded street illumination, compressed shadows, imperfect auto-exposure, no cinematic lighting, no dramatic color grading. Realistic smartphone optics with small-sensor depth behavior, softly compressed background, uneven sharpness, localized softness, slight denoise, mild compression artifacts, imperfect HDR reconstruction, natural skin texture, real hair softness, non-uniform detail recovery, and ordinary mobile-photo visual entropy.
Avoid cinematic composition, luxury editorial polish, influencer aesthetics, beauty-filter skin, fake HDR, over-clean rendering, DSLR bokeh, perfect symmetry, over-sharpening, plastic skin, porcelain smoothing, artificial facial reshaping, waist slimming, torso narrowing, hip narrowing, thigh reduction, leg elongation, hourglass enhancement, model-like posture, mannequin hands, fused fingers, broken wrists, floating limbs, disconnected shadows, fake railing contact, impossible street perspective, sterile environment, showroom cleanliness, and any smartphone angle that changes the LoRA reference body proportions.
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”.