Home AI ToolsNano Banana

Nano Banana Prompt Builder

Build structured natural-language image prompts, then compress them for tighter context windows.

Structured Builder

Aspect Ratio

Medium

Style Lineage

Subject Category

Subject

Subject Attributes

Composition

Camera or Viewpoint

Action or Scene Behavior

Environment and Background

Lighting

Color System

Mood

Texture and Finish

Control Notes

Stats: 675 chars natural | 614 chars compressed | 9% savings

Prompt Outputs

4:3, Live-action photography, soft-focus haze with an early digital CCD and vintage direct-flash feel. a young woman with damp fringe, blue contact lenses, a brown-red velvet halter top, and a gray-blue lace choker. tight top-down close-up with the subject filling most of the frame, from an overhead viewpoint with intimate framing. leaning forward and looking up into the lens. a minimal dark background with a soft warm cast shadow behind her. direct frontal flash with soft haze across the skin. a muted retro palette with orange-pink blush accents. restrained, intimate, and slightly innocent. subtle film grain and a matte skin finish. keep the frame minimal and close.

How to use: Start from a preset or build from scratch. The natural prompt is best for direct model input, the compressed prompt is useful for tighter context windows, and the block view helps inspect the visual structure.

How to Use

Choose a preset or fill the structured image fields on the left.
The builder assembles a natural-language prompt from format, subject, composition, lighting, color, and mood layers.
Switch between Natural Prompt, Compressed Prompt, and Block View to inspect the output.
Click Copy to export the version that best fits your image model workflow.

About This Tool

Build structured natural-language image prompts, then compress them for tighter context windows.

This tool processes everything locally in your browser. Your files and data never leave your device. No uploads, no tracking, no accounts required.

Related Tools

Nano Banana Structured Image Prompt Guide

Build image prompts through structured visual layers instead of starting from a blank textarea. Define format, medium, style lineage, subject, composition, lighting, color, mood, and finish, then copy either the natural prompt, a compressed variant, or a labeled block view.

Frequently Asked Questions

Does this only work for portrait prompts?

No. The builder supports portraits, products, interiors, architecture, food, abstract posters, and other scene-based image prompts.

Can I still get a shorter output for small context windows?

Yes. The tool generates a compressed prompt alongside the full natural-language version, so you can choose the format that best fits your model.