What Nano Banana 2.1 does
Nano Banana 2.1 is Google's latest Nano Banana image model. It generates from text and edits from up to 10 reference images, with mask and ink-based targeting for precise local changes, multi-turn character consistency, and sharper text and infographic layouts. Output goes up to 4K, and it costs less per image than Nano Banana 2 — the best default for photo-real edits, product shots and consistent characters.
- Up to 10 Reference Images
- Blend up to 10 inputs into one coherent output — products, people, styles and backgrounds stay consistent across the composition.
- Mask & Ink-Based Editing
- Mark the area to change — a mask, a circle, an arrow — and the edit lands exactly there while the rest of the image stays untouched.
- Character Consistency
- Multi-turn editing keeps the same face, outfit and proportions from one iteration to the next — built for characters, mascots and campaigns.
- Text, Infographics, 4K
- Improved text rendering and infographic layout, with 1K for drafts, 2K for standard work and 4K for print — all at a lower price than Nano Banana 2.
How to use it
Describe or Upload
Type a prompt, or add up to 10 reference images for an edit or a composition.
Point at the Change
For local edits, mark the area or just say what to change — 'swap the jacket', 'add the logo to the mug'.
Pick Resolution & Generate
Choose 1K, 2K or 4K and get the result in seconds. Keep iterating in the same conversation.
Questions
Nano Banana 2.1 is Google's newest Nano Banana image model: text-to-image and image editing with up to 10 references, mask-based local edits, multi-turn character consistency and output up to 4K.
Compared with Nano Banana 2 it takes more references (10), adds mask and ink-based editing, renders text better and costs less per image. Nano Banana Pro remains the Gemini Pro Image tier for the most demanding 4K production work.
1 credit per image at 1K, 2 credits at 2K and 2.5 credits at 4K. The price per tier is shown in the composer before you generate.
Yes. Mask or mark the area — a circle, a bounding box, an arrow — and describe the change. The rest of the image stays as it was.