Lock identity, composition, lighting, and wardrobe before style words so a portrait series stays consistent instead of drifting frame to frame.
El prompt
Create an editorial portrait of [SUBJECT] for [USE]. Preserve facial identity and natural skin texture. Composition: [FRAMING]. Lighting: [LIGHTING]. Wardrobe: [WARDROBE]. Background: [BACKGROUND]. Color grade: [PALETTE]. Avoid plastic skin, distorted hands, text, logos, and over-sharpening.
Reemplaza las variables
[SUBJECT]The person or reference image.
[FRAMING]e.g. head-and-shoulders, 3/4.
[LIGHTING]e.g. soft key, rim light.
Qué modelo usar
- Nano Banana Pro — Identity-preserving image edits are handled by the image model, not the text model.
Entrada de ejemplo
- [SUBJECT] = uploaded reference photo. [USE] = "LinkedIn headshot".
Resultado de ejemplo
Solo publicamos resultados de ejemplo de ejecuciones reales del modelo. Esta plantilla aún no se ha probado formalmente, por lo que no se muestra ningún resultado. Ejecútala tú mismo con la entrada de arriba.
Por qué funciona esta estructura
- Structure-before-style stops adjectives from fighting each other.
- An explicit negative list removes common artefacts.
Fallos comunes y soluciones
Face drifts between generations.
Solución: Reuse the same reference and keep identity terms first in the prompt.
Limitaciones conocidas
- Image models vary; results are not guaranteed and need review.
Preguntas frecuentes
Where are the tested portrait prompts?
The image gallery has portrait prompts with real example outputs. This page is a reusable template for building your own.