This tiny bakery nano banana pro prompt pushes a miniature scene through accurate hands, multiple people and a readable sign.

Morning croissant bakery
A tiny working bakery photographed like an architectural dollhouse cutaway. Three bakers with distinct aprons shape croissants at a flour-dusted marble bench; accurate hands, trays, copper oven, hanging linen, and morning window light. A chalkboard menu must show the exact legible words TODAY: CROISSANTS. Warm cinematic miniature realism, 35mm tilt-shift lens, vertical composition.
Settings: Google Nano Banana Pro · 3:4 · 2K · safety OFF
Variations
Night bakes

A tiny working night bakery photographed like an architectural dollhouse cutaway. Two bakers in blue aprons glaze pastries at a flour-dusted marble bench; accurate hands, brass pendant lights, copper oven, stacked bread crates, rain on the window. A chalkboard menu must show the exact legible words NIGHT BAKES. Moody miniature realism, 35mm tilt-shift lens, vertical composition.
Settings: Google Nano Banana Pro · 3:4 · 2K · safety OFF
Seaside lemon tarts

A tiny seaside bakery photographed like an architectural dollhouse cutaway. Three bakers with natural hands arrange lemon tarts at a flour-dusted marble bench; sea-blue tiles, striped awning, bright window light, copper oven, baskets of lemons. A chalkboard menu must show the exact legible words LEMON TARTS. Fresh miniature realism, 35mm tilt-shift lens, vertical composition.
Settings: Google Nano Banana Pro · 3:4 · 2K · safety OFF
Why this Tiny Bakery Nano Banana Pro prompt works
- The cutaway framing establishes the dollhouse scale at once.
- Separate baking jobs give each person a purpose and reduce visual clutter.
- The sign is a deliberate stress test for text rendering while the hands test detail handling.
More Nano Banana Pro prompts
Browse all Nano Banana Prompts. For another food-image approach, see GPT Image 2 Prompts.
More on the model itself: Nano Banana Pro on Google DeepMind.
Run this prompt yourself
Every example above was generated on Wiro. Paste the prompt, pick the model, and get your own result.
1 Comment