Nano Banana

Nano Banana is Google’s name for a family of image-generation and editing models integrated into Gemini and selected Google products. The original Nano Banana was introduced as a fast, efficient model for everyday visual creation, while newer versions include Nano Banana Pro and Nano Banana 2. Google identifies Nano Banana 2 as Gemini 3.1 Flash Image, combining advanced image capabilities with faster generation.

Users can create images from written prompts or upload existing visuals and describe the changes they want conversationally. The model can replace backgrounds, add or remove objects, modify outfits, change environments, restore photographs, apply visual styles, and transform individual parts of an image while attempting to preserve the original subject. Follow-up prompts allow users to continue refining an image without restarting the project.

Nano Banana supports reference-based generation and multi-image composition. Several uploaded images can be combined into a new scene, while styles, textures, colors, subjects, and compositions can be transferred between references. A major strength of the model is maintaining the recognizable appearance of people, pets, products, or characters across successive edits and generated scenes.

Newer models improve text rendering, infographic creation, diagram generation, real-world knowledge, and the visual representation of specific subjects. Nano Banana 2 can also use information and images from web search to produce more accurate visuals when current or factual context is important.

Nano Banana is available through Gemini’s web and mobile experiences, with model access and generation limits depending on the selected Gemini mode, account, subscription, country, and age. Google applies SynthID watermarking to generated images and advises users to review outputs for copyright, privacy, and factual accuracy before publication.