GPT Image Prompt
EXECUTION PROMPT
Create a casual vertical iPhone-style indoor flash photo of the exact same real woman from the active LoRA identity reference. Preserve her LoRA-defined facial geometry, eye spacing, face shape, nose, lips, jaw, cheek structure, fair skin tone, long black hair identity, natural upper-body proportions, shoulder-to-head relationship, shoulder width, torso presence, body mass impression, and recognizable physical presence. Do not beautify, slim, reshape, smooth, model-ify, or replace her with a similar-looking woman.
She is very close to the camera against a plain ordinary indoor wall, wearing the same black thin-strap camisole with a V neckline and a pale pink knit cardigan slipped loosely off her shoulders. Keep the small faint text mark near her collarbone and keep her long pale nails intact. The clothing should look physically worn, with natural fabric tension around the straps, neckline, shoulders, and cardigan edges, not painted-on or fashion-styled.
Her raised hand stays very close to the lens on the left side of the vertical frame, thumb and index finger pinched slightly apart as if caught mid-gesture. The foreground hand is mildly enlarged by close smartphone wide-angle perspective and softly out of focus, with natural finger spacing, believable knuckles, correct thumb placement, realistic nail alignment, and no fused or duplicated fingers. Her face remains clearer behind the hand, partially visually interrupted by the near-lens gesture in a believable casual phone-photo way.
Her expression remains subtle and real: relaxed lowered eyelids, a soft camera-aware gaze, slightly parted lips, quiet private-moment stillness, mild natural asymmetry, alive eyes, and relaxed mouth tension. Do not copy blank LoRA-reference stiffness, dead eyes, mannequin neutrality, seductive influencer gaze, commercial smile, or cinematic acting emotion.
All expression, gesture, outfit, pose, camera behavior, lighting, framing, and realism details remain subordinate to the active LoRA identity reference. The final subject must remain unmistakably the exact same LoRA woman in face and body presence, not a similar substitute, not a beautified replacement, not a generic attractive version, and not the same face placed on altered upper-body proportions.
Use imperfect handheld vertical iPhone framing with close indoor distance, slight crop imbalance, casual asymmetry, ordinary wall background, harsh direct phone flash, clipped highlights on the foreground hand and nails, flattened flash skin tones, strong cast shadow on the wall, mild red-eye/flash harshness if natural, uneven sharpness, slight denoise, compression softness, local texture instability, and smartphone ISP imperfections. The image should feel like a real private indoor flash photo captured casually, not a polished portrait.
Avoid cinematic lighting, DSLR depth of field, luxury editorial composition, influencer aesthetics, beauty-filter skin, porcelain smoothness, perfect symmetry, over-clean HDR, globally sharp rendering, artificial bokeh, fake portrait-mode separation, over-flattering camera angle, face slimming, body reshaping, narrowed shoulders, altered upper-body proportions, blank stare, frozen mouth, plastic hands, fused fingers, duplicated fingers, impossible pinch gesture, floating clothing, synthetic wall texture, CGI cleanliness, and staged fashion-photo energy.
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”.