Nano Banana 2.1

Introducing Nano Banana 2.1 — better visual design, mask-based editing, subject consistency, and natural images.

Visit website

Google's image generation and editing model with sharper design, mask editing, and 1K–4K output via API.

Model announcement hero

Model announcement hero

1 / 3

Overview

What is Nano Banana 2.1?

Introducing Nano Banana 2.1 — better visual design, mask-based editing, subject consistency, and natural images.

Nano Banana 2.1 is Google's latest image generation and editing model, an update to Nano Banana 2 (Gemini 3.1 Flash Image). Google says it outperforms earlier Nano Banana models across the board, with better visual design, mask-based editing, stronger subject consistency, and more natural-looking images.

It generates 1K images by default, upscalable to 2K and 4K, with conversational mask-based editing, up to 14 reference images per workflow, accurate text rendering for infographics and posters, and Search grounding for fact-checked generation. All generated images include a SynthID digital watermark.

Try it in the Gemini app or Google AI Studio, or build with it via the Gemini Developer API and Google Cloud. API pricing is pay-as-you-go (from $1.50/1M input tokens, about $0.034 per 1K image); this model has no free API tier.

Platforms and languages

Nano Banana 2.1 Availability

Platforms

WebApi

Languages

EnglishArabicGermanSpanishFrenchHindiIndonesianItalianJapaneseKoreanPortugueseRussianUkrainianVietnameseChinese

Capabilities

Nano Banana 2.1 Key Features

High-resolution output

Default 1K image generation, upscalable to sharp 2K and 4K output ready for production use.

Mask-based editing

Define masks conversationally to edit only specific regions of an image, leaving the rest untouched.

Subject consistency

Reference up to 14 images in one workflow, keeping up to 4 characters consistent and 10 objects high-fidelity.

Search grounding

Combines Google Web and Image Search to verify facts and generate images from real-world information.

Accurate text rendering

Renders clean, readable stylized text directly in images — for infographics, menus, and posters.

Best for

Who uses Nano Banana 2.1?

Creators and designers

Generate and edit images in the Gemini app or AI Studio

Developers

Build image generation into apps via the Gemini API

Enterprises

Deploy image generation on Google Cloud / Vertex AI

Plans and access

Nano Banana 2.1 Pricing

Paid

API from $1.50/1M input tokens; ~$0.034/1K image; no free tier; 5K free search requests/mo.

Common questions

Nano Banana 2.1 FAQs

Is Nano Banana 2.1 free?

No. The API has no free tier for this model; it is pay-as-you-go. Gemini app / AI Studio pricing is not published on the model page.

Where can I use it?

In the Gemini app, Google AI Studio (login required), the Gemini Developer API, Google Cloud / Vertex AI, and the Enterprise Agent Platform.

What resolutions does it support?

1K by default, with 2K and 4K output supported.

Do generated images have a watermark?

Yes. All generated images include a SynthID digital watermark.

How is my data handled?

The model page only documents the SynthID watermark; data retention and training-use details are not published there.

Reviews

See what the community thinks and share your experience.

—

Based on 0 ratings

Rating distribution

0
0
0
0
0

Leave a review

Sign in to rate

Community reviews

No reviews

No written reviews yet.

Explore by category

AI Image Generation

View all AI Image Generation websites
Mareel preview
Mareel

Mareel is an all-in-one AI video and image platform that brings 40+ leading AI models into one workspace.

AI Image Generation

Explore by category

AI Creation Generation

View all AI Creation Generation websites
Framer preview
Framer

Framer's AI agents design, update and improve websites from a text description — styling, layout, CMS and SEO included.

AI Creation Generation

Explore by category

AI Development Infrastructure

View all AI Development Infrastructure websites