What is Nano Banana 2?

Nano Banana 2 is our implementation of Google's Gemini 2.5 Flash Image model — a state-of-the-art multimodal AI that generates and edits images from text and image inputs. It supports text-to-image generation, image-to-image transformation, and targeted image editing.

How many images can I generate per prompt?

Nano Banana 2 can generate up to 10 images per prompt. You can also provide up to 3 reference images as input for image-to-image and editing workflows.

What aspect ratios are supported?

Nano Banana 2 supports 9 aspect ratios: 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, and 16:9 — covering square, portrait, landscape, and widescreen formats.

Are generated images watermarked?

Yes. All images generated or edited by Nano Banana 2 are automatically watermarked with Google's SynthID technology, which embeds an invisible digital watermark and Content Credentials (C2PA) for responsible AI content verification.

How does character consistency work?

Nano Banana 2 can maintain the visual identity of characters, objects, and brand assets across multiple generations. Upload a reference image of your character and describe the new scene — the model preserves appearance, clothing, and style while placing them in the new context.

How many credits does image generation cost?

Each image generation with Nano Banana 2 costs 144 credits, the same shared image generation cost used across the platform.

Nano Banana 2 · Gemini 3.1 Flash Image

Nano Banana 2 AI
Image Generator

State-of-the-art image generation and editing powered by Gemini 2.5 Flash Image. Generate, transform, and edit images with natural language — character consistency, multi-image fusion, and native world knowledge built in.

Generate Free Image to Image

Max Resolution

Aspect Ratios

Ref Images

<2s

Gen Speed

Text to ImageImage to ImageImage EditingBackground RemovalObject RemovalStyle TransferCharacter ConsistencyMulti-Image FusionPhoto ColorizationProduct MockupsRoom RestylingDiagram to IllustrationGoogle Search GroundingImage Search Grounding4K Resolution14 Aspect RatiosSynthID WatermarkThinking ModesMulti-turn Chat

Model IDgemini-2.5-flash-image

Max Input32,768 tokens

Images per promptup to 10

Input imagesup to 3

Aspect ratios9 supported

WatermarkSynthID C2PA

What it can do

Core Capabilities

Nano Banana 2 combines Google DeepMind's multimodal understanding with high-quality image synthesis — four breakthrough capabilities in one model.

Character Consistency

Maintain the exact appearance of characters, objects, and brand assets across multiple prompts and edits. Place the same character in entirely different environments without losing visual identity.

Text-to-Image Generation

Generate up to 10 high-quality images per prompt across 9 aspect ratios (1:1, 16:9, 9:16, 3:2, 4:3, and more). Every output is watermarked with SynthID for content authenticity.

Quick start

Up to 4K resolution · 14 aspect ratios · SynthID watermarked

Try Text to Image

See it in action

Real Examples from Gemini

These are actual output examples from Google's official showcase — demonstrating the three core capabilities of the model.

Fine-grained control over lighting, angles, and focus. Edit any detail in an existing image with a single natural-language instruction.

6 examples · Images sourced from Google Gemini

All 14 supported ratios

Every Format. Every Canvas.

From ultra-wide cinematic 21:9 to extreme vertical 1:8 — no crop, no compromise. Hover each box to see the ratio.

1:1

3:2

2:3

3:4

4:3

4:5

5:4

9:16

16:9

21:9

1:4

4:1

1:8

8:1

1:1·Square3:2·Landscape2:3·Portrait3:4·Portrait4:3·Landscape4:5·Portrait5:4·Landscape9:16·Vertical16:9·Widescreen21:9·Ultrawide1:4·Tall Banner4:1·Wide Banner1:8·Ultra Tall8:1·Ultra Wide

Intelligence level

Thinking Mode Control

Choose between speed and quality. Minimal mode for rapid iteration, High mode for maximum fidelity on production-ready outputs.

Thinking Mode:

~1sLow compute budget

Optimized for speed. Great for rapid iteration, drafts, and high-volume workflows where generation time matters most.

Best for

Rapid prototyping
Batch generation
Quick iterations
High-volume tasks

SpeedQuality

Unique capability

Real-Time Web Intelligence

The only image model with built-in Google Search and Image Search grounding. Generate images anchored to real, current information.

Google Search Grounding

Uniquely powered by real-time Google Search and Image Search — a capability no other image model has.

Live world knowledge

Generate images of real events, places, products, and trending topics using current information from Google Search.

Contextual accuracy

Prompts referencing real-world concepts get grounded in accurate, up-to-date data — not training cutoffs.

Differentiators

What Makes Nano Banana 2 Different

Low Latency & Cost-Effective

Built on the Gemini 2.5 Flash architecture, Nano Banana 2 delivers fast generation at a fraction of the cost of comparable models — ideal for high-volume workflows and iterative creative work.

SynthID Digital Watermark

Every generated and edited image is automatically watermarked with Google's SynthID technology, providing verifiable content credentials (C2PA) for responsible AI use.

Complex Instruction Handling

Unlike models that require multiple steps, Nano Banana 2 can execute complex editing instructions in a single pass — reading diagrams, understanding spatial relationships, and applying multi-part changes at once.

Multimodal Input & Output

Accepts both text and images as input, and outputs text alongside images. This enables conversational image editing workflows where you can describe changes and receive both the edited image and an explanation.

Conversational workflow

Iterate Through Conversation

Unlike one-shot generators, Nano Banana 2 supports multi-turn image iteration. Describe your changes, see results, and refine — in a chat-like loop.

1Start with a text prompt or upload an image
2Ask follow-up changes in natural language
3Refine composition, style, and details iteratively
4Export any step in the conversation as final output

A minimalist poster of a golden banana floating in space

[Image generated] Deep space background, warm golden banana with subtle glow...

Make the background more vibrant with nebula colors and add text "NANO" in the corner

[Updated image] Vivid nebula in purple and teal, bold "NANO" typography applied...

Why Use Nano Banana 2

Generate production-ready images in seconds — no design software required.

Maintain brand and character consistency across entire campaigns with a single model.

Edit photos with natural language instead of complex layer-based tools.

Fuse multiple reference images into cohesive, photorealistic compositions.

Every output includes SynthID watermark for responsible, verifiable AI content.

Cost-effective at scale — ideal for agencies, e-commerce, and content teams.

Applications

Use Cases

E-Commerce & Product Mockups

Generate product lifestyle shots, swap backgrounds, and create consistent product imagery across multiple scenes without a photo studio.

Real Estate & Interior Design

Restyle rooms with different furniture, color schemes, or textures. Generate listing cards and visualize renovation concepts from reference photos.

Brand & Marketing Assets

Create consistent brand characters, mascots, and visual assets across campaigns. Generate posters, social media content, and ad creatives at scale.

Interactive Education

Turn hand-drawn diagrams into polished illustrations. Generate visual explanations, infographics, and educational materials from text descriptions.

Photo Editing & Retouching

Remove unwanted objects, change backgrounds, alter lighting, colorize black-and-white photos, and apply style transfers — all with natural language prompts.

Character & IP Design

Design consistent characters and IP assets, then place them in different environments, poses, and scenarios while maintaining visual identity.

Getting started

How to Use Nano Banana 2

Step 1: Choose your mode

Select Text to Image for pure generation, Image to Image to transform existing photos, or Image Editing for precise targeted changes.

Step 2: Upload references (optional)

For Image to Image or Editing modes, upload up to 3 reference images. Nano Banana 2 will understand and incorporate all of them.

Step 3: Write your prompt

Describe what you want in natural language. Be specific about style, mood, colors, and changes. Complex multi-part instructions work in a single prompt.

Step 4: Generate & refine

Get up to 10 image variations per generation. Select the best result and continue refining with follow-up prompts or download your final assets.

Model Comparison

Nano Banana 2 vs NB Pro

NB2 is fast and cost-effective. NB Pro adds 4K, extreme ratios, and expanded reference support.

Nano Banana 2Standard

512px option
9 core ratios
Up to 3 input images
$0.039/image

NB ProRecommended

Up to 4K resolution
14 aspect ratios
Up to 10 reference images
Thinking mode control

Pricing

Transparent, predictable costs

Standard

$0.039/ image

Immediate generation

Full quality output

SynthID watermark included

Batch

$0.0195/ image

50% discount on batch jobs

Same quality output

Ideal for high-volume workflows

Save 50%

Supported Workflows

Text to ImageImage to ImageImage EditingBackground RemovalObject RemovalStyle TransferCharacter ConsistencyMulti-Image FusionPhoto ColorizationProduct MockupsRoom RestylingDiagram to Illustration

SynthID Watermarking

Every image generated or edited by Nano Banana 2 is automatically embedded with Google's SynthID invisible digital watermark and C2PA Content Credentials — enabling verifiable, responsible AI content at every step.

Invisible watermarkC2PA credentialsTamper-resistantIndustry standard

Common questions

Frequently Asked Questions

Ready to create

Start Creating with Nano Banana 2

From text to 4K image in under 2 seconds. 14 aspect ratios. Google Search grounding. Start free.

Generate Free View pricing

Iterate Through Conversation

Unlike one-shot generators, Nano Banana 2 supports multi-turn image iteration. Describe your changes, see results, and refine — in a chat-like loop.

1Start with a text prompt or upload an image

2Ask follow-up changes in natural language

3Refine composition, style, and details iteratively

4Export any step in the conversation as final output

Nano Banana 2 AIImage Generator