Generate and edit from natural-language prompts
Describe the image you want, generate a first result, then ask for specific changes. Move from an idea to a refined visual without rebuilding the concept from scratch.
Nano Banana 2 online
Nano Banana 2 AI Image Generator & Editor
Generate and edit AI images with precise text, subject consistency, localization, flexible aspect ratios, and up to 4K output.
Model overview
What Is Nano Banana 2?
Nano Banana 2 is Google DeepMind's image generation and editing model, also identified as Gemini 3.1 Flash Image. It is built for fast image creation and editing while bringing advanced image capabilities to a Flash-class model.
The model can generate images from prompts, edit existing visuals, render text inside images, localize image content, keep subjects more consistent across a workflow, and upscale images to 2K or 4K.
This image model is useful for creators, marketing teams, e-commerce brands, designers, and teams that need visual concepts, campaign assets, infographics, posters, or repeatable characters.
Features
Features for Faster Visual Creation
A quick look at what Nano Banana 2 supports before you start a generation.
Create clearer text inside images
Generate posters, greeting cards, marketing mockups, and other visuals that need words inside the image. Review spelling and fine text details before publishing.
Use real-world knowledge for visuals
Create specific subjects, infographics, diagrams, and visual explanations with help from Gemini's real-world knowledge.
Localize images for different markets
Translate text inside images and adapt the visual context for different audiences. Review grammar, spelling, and cultural nuance before use.
Keep characters and objects more consistent
Build stories, campaigns, and multi-scene concepts with stronger subject consistency — up to five characters and fourteen objects across a single workflow.
Resize and upscale for production needs
Move between square, portrait, and landscape layouts, and upscale to 2K or 4K for sharper social posts, ads, presentations, and larger creative assets.
Examples
Image Generation and Editing Examples
Start from a proven scene. Reuse the prompt, change the subject or setting, and generate a new version.
How to use
How to Use Nano Banana 2
Start with a clear prompt, add reference images when you need control, then generate an image you can review and refine.
- 01
Describe the image you want
Write a clear prompt with the subject, setting, composition, text, style, and intended use. More specific instructions give the model more context.
- 02
Add reference images when needed
Upload images that show the character, object, visual style, or source material you want to preserve or reinterpret.
- 03
Choose the layout and output settings
Select a square, portrait, or landscape aspect ratio. If 2K or 4K upscaling is available, choose the output level that fits your final asset.
- 04
Generate, review, and refine
Generate the image, check text and fine details, then use follow-up edit instructions to adjust composition, subjects, colors, or other elements.
Use cases
Built for Real Creative Workflows
Create images for the places where your audience already looks: social feeds, campaigns, product pages, and internal presentations.
Content Creators
Create Instagram posts, TikTok graphics, YouTube thumbnails, storyboards, illustrated sequences, posters, and personal creative concepts. Use text-to-image generation for new ideas, reference images for recurring subjects, and different aspect ratios to adapt one visual direction across social channels.
Create this image
Marketing & Ad Teams
Develop campaign concepts, ad creatives, marketing mockups, posters, localized graphics, product launch visuals, and presentation-ready ideas. Precise text rendering and image localization can help teams explore creative directions for different markets before final design review.
Create this image
E-commerce & Brands
Create product scene concepts, promotional graphics, lifestyle imagery, seasonal campaign assets, and branded visual ideas. Use prompt-based image editing to test backgrounds, layouts, and creative directions while keeping important subjects recognizable across related assets.
Create this image
Enterprise & Industry
Turn notes and structured information into diagrams, infographics, visual explainers, training concepts, and presentation graphics. Gemini's real-world knowledge can help with specific subjects, but factual labels, data, and technical details should be checked before external use.
Create this imagePricing
Nano Banana 2 Pricing & Access
PWZ pricing, plan limits, and model credit usage require verification before publication. Add your verified billing rules here before publishing.
Pricing fieldStatus
Credits per generationNeeds verification
Credits per editNeeds verification
2K / 4K costNeeds verification
Included plan creditsNeeds verification
Free usage or trialNeeds verification
Compare
Nano Banana 2 vs Nano Banana Pro
Nano Banana 2 emphasizes rapid generation, precise instruction following, and image-search grounding. Nano Banana Pro is positioned for high-fidelity work that needs maximum factual accuracy.
NeedBest fit
Rapid image generation and editingNano Banana 2
Precise instruction followingNano Banana 2
Integrated image-search groundingNano Banana 2
Maximum fidelity and factual accuracyNano Banana Pro
FAQ
Nano Banana 2 FAQ
What is Nano Banana 2?
Nano Banana 2 is Google DeepMind's image generation and editing model, officially identified as Gemini 3.1 Flash Image. It supports image creation, editing, text rendering, localization, subject consistency, and higher-resolution workflows.
Is Nano Banana 2 the same as Gemini 3.1 Flash Image?
Yes. Google DeepMind identifies the model as Gemini 3.1 Flash Image. "Nano Banana 2" is the official user-facing name; NanoBanana 2, Nano Banana2, and Gemini Nano Banana 2 are common search variations, not separate models.
Can Nano Banana 2 generate and edit images?
Yes. The model is designed for both AI image generation and image editing. You can create an image from a prompt and refine existing or generated visuals with additional instructions.
Can Nano Banana 2 generate text inside images?
Yes, but you should review the result. Google highlights precise text rendering as a capability, while its model page also notes that spelling, small details, and long text can still contain errors.
Can Nano Banana 2 create 4K images?
Google documents 2K and 4K upscaling on the official model page. The exact output settings available through PWZ depend on the implemented interface.
Can Nano Banana 2 keep characters consistent?
Yes, with limitations. Google states that the model can preserve the resemblance of up to five characters and the fidelity of up to fourteen objects across a single workflow, though consistency is not guaranteed in every result.
Can Nano Banana 2 translate or localize images?
Yes. The model can translate text and localize content inside images. Google also warns that grammar, spelling, idioms, and cultural nuance may need human review.
Is Nano Banana 2 free on PWZ?
PWZ pricing, free usage, plan limits, and credit costs require verification. Publish only the actual access terms shown in your billing system.
Can I use Nano Banana 2 images commercially?
Commercial-use rights depend on the terms that apply to your PWZ service and the underlying model. Do not publish a commercial-use promise until the applicable terms have been reviewed.
What image formats does PWZ support for Nano Banana 2?
The provided official model information does not define PWZ upload or export formats. Add the exact supported formats and file-size limits from your production implementation.
Do I need to install Nano Banana 2 or use a local GPU?
No local model installation is required when you use the PWZ web interface. You create and edit images through the online tool; browser requirements and account limits should be confirmed in the production implementation.
Nano Banana 2 vs Nano Banana Pro: what is the difference?
Google positions Nano Banana Pro for high-fidelity work that needs maximum factual accuracy, while Nano Banana 2 emphasizes rapid generation, precise instruction following, and integrated image-search grounding. Choose based on the workflow rather than assuming one model is better for every task.
Related pages
Continue exploring
Ready to create
Create Your Next Image with Nano Banana 2
Generate a new visual, edit an existing concept, add text, localize creative assets, or build a consistent image sequence with the model on PWZ.
Start Creating