GPT Image Prompt
Photorealistic casual iPhone-style execution prompt: preserve the active LoRA identity reference as the absolute subject identity, including the exact same recognizable oval face geometry, compact jaw and chin, pale skin tone, large light eyes without enlarging or beautifying them, the same pale blonde straight bangs and high side ponytail, and the same compact full-body proportions and physical presence. Preserve LoRA body geometry exactly: shoulder-to-head relationship, compact torso-to-leg balance, natural waist thickness, hip width, thigh volume, leg length ratio, body mass distribution, grounded adult height impression, and the same non-model-like physical presence. Do not slim, elongate, narrow, curve-enhance, or replace the body with a generic attractive silhouette.
The LoRA subject stands inside a cramped urban glass phone booth, physically grounded in the tight interior space. She holds a lime-green payphone receiver against her left ear, with the cord visibly connecting to the lime-green payphone hardware. Her other arm reaches forward toward the camera, palm open and close to the lens; the foreground fingers are naturally larger from close smartphone wide-angle distance, slightly soft, with believable finger count, natural knuckle structure, realistic palm compression, and no fused or duplicated fingers. Her stance feels casual and weight-bearing, not posed like a fashion model, with believable balance inside the limited booth space.
She wears an off-shoulder black long-sleeve crop top with a colorful graphic print, very short denim shorts with a belt and worn edges, a bright beaded necklace and bracelets, black open-toe platform shoes, and thick fuzzy striped leg warmers. Clothing follows gravity and body contact naturally, with real fabric tension around the shoulders, waist, hips, shorts, sleeves, and leg warmers. The outfit must not reshape the LoRA body identity or make the waist thinner, legs longer, hips narrower, or thighs smaller than the reference.
Her expression is subtle and alive, not copied from any stiff reference frame: a soft direct glance toward the phone camera, relaxed eyelids, natural eye reflections, and a small softly parted mouth as if caught mid-interaction while reaching forward. Preserve the source-prompt human moment while keeping the LoRA facial structure unmistakable. Avoid blank stare, dead eyes, mannequin neutrality, seductive influencer gaze, beauty-commercial smile, or exaggerated cinematic acting.
The booth interior is tight, ordinary, and physically real: glass walls with slight smudges and reflections, a ceiling light, lime-green payphone hardware, visible cord, narrow booth framing, practical signage outside through tinted glass, bright city colors beyond the booth, and slightly awkward real-world object spacing. Maintain believable environmental scale, realistic booth geometry, coherent reflections, natural contact shadows, and physically plausible subject-to-booth proportions. The subject exists inside the booth, not composited over it.
All expression, gesture, outfit, pose, camera behavior, lighting, and realism details remain subordinate to the active LoRA identity reference. The final subject must still be unmistakably the exact same LoRA person in both 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 different body proportions.
Camera style: close handheld iPhone-style smartphone snapshot from slightly above and close in front of the booth, casual human-operated framing, mild wide-angle perspective, slightly imperfect horizon, awkward crop balance, subtle framing asymmetry, and believable near-hand enlargement without distorting the LoRA body identity. The image should feel captured spontaneously by someone standing close, not directed by a professional photographer.
Optics and processing: realistic small-sensor smartphone depth behavior, uneven sharpness between the face and near hand, near-hand softness from proximity, slight edge stretching, mild geometric imperfection, practical auto-exposure, localized compression artifacts, subtle HDR reconstruction instability, non-uniform skin and fabric detail, natural smartphone denoise, slight glass glare, and imperfect texture recovery. Use mixed practical daylight from outside signage and overhead booth lighting, with uneven illumination, mild highlight clipping on glass/signage, realistic shadow compression, and no cinematic lighting polish.
Realism priorities: preserve LoRA face and body identity, believable contact physics, grounded stance, realistic hand anatomy, natural skin texture, physical booth scale, smartphone imperfection, visual entropy, and socially believable casual capture. Avoid cinematic composition, luxury editorial aesthetics, influencer beauty optimization, polished fashion-photo energy, artificial HDR perfection, DSLR bokeh, over-clean skin, plastic texture, fake phone-booth geometry, floating feet, disconnected shadows, mannequin posture, over-centered composition, over-sharpening, symmetrical AI perfection, or any body beautification that changes the reference 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”.