A structured JSON template designed for generating a multi-frame editorial fashion portrait series (grid style). It outlines common elements like lighting (direct flash/golden hour, high contrast) and realism, and provides customizable slots for two distinct frames, allowing the user to define specific poses, angles, and visual prompts for each section of the grid while maintaining facial consistency.
{
"meta": {
"purpose": "Editorial fashion portrait series (multi-frame grid style, high virality)",
"version": "Nano Banana Pro optimized",
"aspect_ratio": "16:9 or 1:1 grid"
},
"common_elements": {
"subject": "[e.g., Young woman with {argument name=\"hair description\" default=\"long dark hair\"}, {argument name=\"skin tone\" default=\"fair skin\"}]",
"face_constraint": "Preserve original face structure and proportions exactly from reference",
"lighting_and_style": "Direct flash or golden hour, high contrast, vintage/cinematic aesthetic, saturated colors, hyper-realistic textures, 8K"
},
"frames": [
{
"frame_id": 1,
"position": "{argument name=\"frame 1 position\" default=\"Top Left\"}",
"visual_prompt": "{argument name=\"frame 1 visual prompt\" default=\"Low-angle shot, subject perched on object, leg extended dramatically toward camera, sharp foreground texture\"}",
"pose_details": "[e.g., Leaning back, looking away casually]"
},
{
"frame_id": 2,
"position": "[e.g., Top Right]",
"visual_prompt": "[e.g., High-angle shot looking down, hand on prop, wide lens distortion]",
"pose_details": "[e.g., Seated, head tilted up]"
}
],
"negative": ["blurry", "plastic skin", "distorted face", "low res", "cartoon", "artifacts"]
}CC BY 4.0. Prompt pages keep the original attribution and source links from the upstream gallery records stored in the database.


