
Document original
GPT Image 2 · en
Scene:
A square portrait staged in a dark chamber lit by candles, embers, or a single hidden lantern, with heavy shadows, warm ochre skin tones, black velvet darkness, and a painterly yet physically believable material presence. The frame should feel ceremonial, intimate, and heavy with air, fabric, and oil-rich color. Use a judicial chamber with a dark oak chair, velvet curtain, and two faint candles low in the frame.
Subject:
The subject is a fantastical humanoid beast presented with aristocratic gravity: horned, feathered, scaled, furred, or antlered, dressed in ornate garments and posed as if for an old court portrait. The creature should feel specific, dignified, and psychologically readable rather than monstrous chaos. A stag-antlered magistrate in black robe and white collar sits upright with hands folded, presenting judgment without motion.
Important details:
The visual law is baroque portrait discipline fused with invented natural history. Preserve luxurious fabric folds, worn metal embroidery, candle smoke, skin texture, fur detail, and solemn hand placement. Avoid cartoon exaggeration, playful fantasy tropes, or bright modern polish. The antlers should create a formal crown-like silhouette, but the emotional tone must remain grave and human.
Use case:
Baroque Bestiary Candlelight single study 5/9.
Constraints:
No readable text, no watermark, no logos, preserve exact 1:1 aspect ratio, no crowns copied from real regalia, no readable heraldic text, no slapstick fantasy, no modern makeup styling, no glossy digital sheen, no broken anatomy, no low-detail background, no accidental text fragments.Mode d’emploi
- Copiez le prompt original complet, paramètres compris.
- Adaptez le sujet, le texte et la composition à vos besoins.
- Utilisez un outil externe compatible ; vérifiez l’accès et le résultat.
Avant de commencer
- Vérifiez si une image de référence est nécessaire. La compatibilité ne prouve pas le modèle utilisé pour cet exemple.
- Vérifiez noms, chiffres et textes ; les versions du modèle peuvent produire des résultats différents.