Image creation
Nano Banana 2.1
Google’s faster, cheaper image model with sharper editing.
Image creation
Google’s faster, cheaper image model with sharper editing.
Product insights
Better edits and character consistency than Nano Banana 2 at roughly half the API price.
Multi-step image edits often drift, changing faces, products, or layouts you wanted to keep.
Designers, marketers, content creators, and developers building image features.
Free to try in the Gemini app. API pricing is about $0.034 per 1K image and $0.076 per 4K image, with a further 50% off in batch mode.
Nano Banana 2.1 is the latest version of Google’s Nano Banana image model. It generates and edits images from text and up to 14 reference images, keeps up to four characters and ten objects consistent, supports aspect ratios as wide as 8:1, and can ground results in Google Search. It is available in the Gemini app, Google AI Studio, Flow, Stitch, and the Gemini API.
Use the Gemini app for quick edits, or Google AI Studio to test prompts and settings.
Describe the image and upload up to 14 reference images for characters, products, or style.
Refine areas with follow-up edits, then export up to 4K or call the model from the API.
Sharper mask-based editing, better subject consistency across edits, adjustable thinking time, and roughly 50% lower API prices than Nano Banana 2.
In the Gemini app, Google Search AI Mode, Google AI Studio, Flow, Stitch, and through the Gemini API.
You can try it free in the Gemini app with daily limits. API use is paid per image.
Official website snapshot
aistudio.google.com
Nano Banana 2.1 is Google’s image generation and editing model, released on October 6, 2026. It improves mask-based editing and subject consistency over Nano Banana 2, outputs up to 4K, and costs about half as much per image through the Gemini API.