Google is rolling out a more powerful version of its Nano Banana AI image model for free users. Nano Banana 2 (also known as Gemini 3.1 Flash Image) is launching today on the Gemini app and other Google AI platforms, making available to everyone the knowledge and rendering features that were previously exclusive to Nano Banana Pro. Google says the update aims to “bring the high-speed intelligence of Gemini Flash to visual generation”, making complex images faster, cheaper, and easier to generate.
Like the Nano Banana Pro, the Nano Banana 2 model uses real-time information, web search images, and Gemini’s real-world knowledge base. Google DeepMind product manager Naina Raisinghani says this provides more relevant data for creating infographics or diagrams, and allows Nano Banana 2 to more accurately represent “specific topics”, although examples of such topics were not provided.
Other features inherited from Nano Banana Pro include the ability to generate images with accurate, legible text and local translation. These capabilities previously required a paid subscription to Google AI Plus, Pro, or Ultra to access in Gemini, but will now be extended to free Gemini users and AI modes in Google Search.
The Nano Banana 2 also offers more creative control over the images generated than the original Nano Banana model. Google says visual improvements include the ability to more rigorously follow complex image requests, as well as more vibrant lighting, richer textures, and sharper details. The appearance of up to five characters and 14 objects can be maintained more consistently in a single workflow, and users get “full control” over aspect ratios and image resolutions ranging from 512px up to 4K.
The new Nano Banana 2 model will replace the Nano Banana Pro option in the Fast, Thinking and Pro generation modes of the Gemini app. Google says AI Pro and Ultra subscribers will still be able to access Nano Banana Pro “for specific tasks” by selecting the three-dot menu on images to revive images. The new model is also rolling out to Google’s AI video tool, Flow, along with a new default image generation model for AI modes in Search, Google Lens, Google apps, and browsers on mobile and desktop.
