Documento original
Nano Banana Pro · en
{
"project_title": "Miniature Coffee Dream",
"scene": {
"setting": {
"location": "Cozy Vintage Coffee Shop Diorama",
"lighting": "Warm golden ambient light with soft bokeh",
"mood": "Comfortable, nostalgic, handcrafted miniature world"
},
"main_subject": {
"character": {
"type": "ultra-detailed 3D chibi girl",
"face": "same as reference face, recognizable features and expression",
"appearance": "large chibi head, small body proportions, smooth fair skin, expressive eyes",
"hair": "same hairstyle as reference",
"expression": "bright, curious smile looking upward",
"pose": "sitting on a tiny wooden stool while watching the coffee drip",
"outfit": "oversized soft white shirt with realistic cotton texture"
},
"object": {
"item": "mini cold brew coffee tower",
"materials": "glass containers, brass fittings, wooden stand",
"details": "spiral glass tubes, dripper, collection flask and ice reservoir",
"moment": "single drop of coffee falling from the dripper captured mid-air"
}
},
"environment": {
"details": [
"rustic wooden counter scattered with coffee beans",
"tiny ceramic cup with spoon",
"background shelves with jars of coffee beans",
"small vintage books and old brewing tools",
"warm fairy lights creating soft glowing bokeh"
]
},
"composition": {
"aspect_ratio": "9:16 vertical",
"focus": "sharp focus on the chibi girl's face and the falling coffee drop",
"depth_of_field": "shallow background blur",
"style": "miniature handcrafted diorama with photorealistic materials"
},
"technical": {
"render": "ultra detailed 3D render",
"textures": "realistic skin pores, fabric weave, aged wood grain, reflective glass and liquid clarity",
"camera": "cinematic close-up feel, 35mm lens, f/1.8"
}
}
}Cómo usar
- Copia el prompt original completo, incluidos los parámetros.
- Adapta el tema, el texto y la composición a tus necesidades.
- Usa una herramienta externa compatible; comprueba el acceso y el resultado.
Antes de empezar
- Comprueba si necesita una imagen de referencia. La compatibilidad no demuestra qué modelo generó el ejemplo.
- Revisa nombres, cifras y textos; el resultado puede variar entre versiones del modelo.
