Nano Banana 2.1

Nano Banana 2.1 is Google's image generation and editing model, identified by the developer model ID gemini-nano-banana-2.1 and described in Google's own documentation as a refined step in the Nano Banana line. The release is framed as the most advanced image model in that family to date, with stated gains in reasoning, visual quality, character and subject consistency, creative controls, in-image text rendering, and 2K resolution output. It is positioned for creators and product builders who need a single workflow for ideation, generation, and iterative refinement rather than a general-purpose chat model.

The model's most distinctive interface choice is instruction-driven editing: a single reference image is passed alongside a natural-language prompt, and it returns the same image with the requested change applied, while the model figures out where the change should land. Documented upgrades include mask-based editing, maintenance of a character's appearance across multiple generated images, finer control over lighting, camera angles, and aspect ratios, and the ability to combine two uploaded pictures into a single composite. Practical fits include branded assets and invitations, stylized figure renders, retro look-and-alib style retouching, and multilingual in-image text, where the model aims to keep subjects registered and text legible through repeated edits.

Eden AIgoogle/gemini-nano-banana-2.1gemini-flash

Open provider page Provider docs

Quick Info

Provider: Eden AI
Model key: google/gemini-nano-banana-2.1
Release date: Oct 6, 2026
Last updated: Oct 6, 2026
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $1.50
Output token cost: $7.50

Limits

Output tokens: 32,768 tokens
Context window: 65,536 tokens

Transparent token rates

Compare Nano Banana 2.1 pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Nano Banana 2.1

Kilo Gateway

Coverage

Google launched Nano Banana 2.1, a new image generation and editing model in the Gemini 3 family that the company says produces better-looking visuals, follows detailed prompts more accurately, and handles edits more reliably than its predecessor. According to Google, the model can generate images for nearly half the price of the prior version, lowering cost for developers. The article states Nano Banana 2.1 is based on Google's Gemini 3.6 Flash model and is rolling out across the Gemini app, Google AI Studio, Gemini API, Search AI Mode, Google Ads, Flow, and Stitch. Improvements target instruction following and consistency of key elements during edits, broadening the model's reach across consumer and enterprise Google surfaces.

Kilo Gateway

Coverage

Google unveiled Nano Banana 2.1 on October 6, 2026, via Google AI Studio, positioning it as its most advanced image model with improved reasoning, quality, and visual design. The model introduces stronger subject and character consistency, mask-based editing, and 2K-resolution output, enabling more reliable image generation and targeted modifications across creative workflows. New creative controls let users adjust camera angles, focus, texture, color, and style from reference photos, and combine two uploaded images into one composition via improved context awareness. Upgrades also cover clearer multilingual text rendering for logos, posters, and comics, plus avatar and figurine-style image features for personalized content inside Gemini.