Video Enhancer
AI Image Model

Nano Banana Pro

Studio-grade images that get text, facts and faces right

Make infographics, posters, localized designs and multi-reference scenes from one prompt, then deliver at up to 4K.

Creative workspace

Nano Banana Pro

27

What is Nano Banana Pro?

Nano Banana Pro is Google DeepMind's Gemini 3 Pro Image model, released in November 2025. It generates and edits images using Gemini's reasoning and world knowledge, renders legible text in many styles, combines up to 14 reference images while keeping up to five people consistent, and outputs at 1K, 2K or 4K.

Nano Banana Pro at a glance

Output
1K / 2K / 4K
References
Up to 14 images
Consistency
Up to 5 people

What you can make with the Nano Banana Pro tool

These are the jobs people actually bring to Nano Banana Pro, and the settings each one needs.

01

Infographics and diagrams

Recipes, how-tos and explainers where both the facts and the labels need to be correct.

02

Localized campaigns

Translate the text on packaging, signs and ads into other languages while the rest of the design stays the same.

03

Composites from many references

Bring characters, products and props from separate images into one scene that looks shot together.

Nano Banana Pro

What Nano Banana Pro does well

Every image below is one of Google DeepMind's published Nano Banana Pro examples. Each one shows a capability the model is built around.

Text you can readOfficial example · Google DeepMind

01

Text you can read

Whole sentences come out spelled correctly, even when the letters are built from objects in the scene, like logs stacked by a woodchuck.

Grounded in real-world knowledgeOfficial example · Google DeepMind

02

Grounded in real-world knowledge

Gemini's knowledge fills in the facts, so a step-by-step recipe or a science diagram comes out with the right ingredients and labels.

Translate and localizeOfficial example · Google DeepMind

03

Translate and localize

Swap the language on packaging or an ad and keep layout, style and product intact.

Many references, one sceneOfficial example · Google DeepMind

04

Many references, one scene

Up to 14 input images can be combined, keeping each character or object recognizable in the result.

Nano Banana Pro

Choose with context

Where Nano Banana Pro is the right call, and where a different model does the job better.

Compared with Nano Banana 2

Pro reasons more before drawing; Nano Banana 2 is faster and cheaper per image

Compared with GPT Image 2

Strong on factual infographics, localization and multi-reference composites

Best decision signal

The image has to be correct, not only attractive, and may need 4K

How to prompt Nano Banana Pro

01

Add references if you have them

Upload up to 14 images of the people, products or styles to keep, or start from text alone.

02

Write the brief

Describe subject, layout and style, put any words that must appear in quotation marks, and name the language for translated text.

03

Pick size and resolution

Choose the aspect ratio and 1K, 2K or 4K, then generate. Results appear in Studio.

Nano Banana Pro questions

Start Creating With Nano Banana Pro

Set up the brief now, then continue in the shared Studio workspace.