Google has announced the launch of Nano Banana 2, the latest evolution of its image generation model that promises to redefine the concept of speed and creative quality. Also known as Gemini 3.1 Flash Image, the new model represents a fundamental turning point thanks to its ability to condense advanced features typical of the Pro version into the lightning-fast execution speed of the Flash line.
Nano Banana 2 is official: the new era of Google’s image generation begins
As the previous Pro model (launched last November) Nano Banana 2 can process extremely high-quality images while adding unprecedented speed (borrowed from the Flash model), enabling an exceptionally smooth editing and iteration process.
Thanks to the integration with Gemini’s vast knowledge base and to real-time information from web research, the model is able to generate infographics, complex diagrams and data visualizations with maximum coherence.
One of the most significant innovations for creators concerns managing the subject consistency, a feature that allows maintaining the likeness of up to five characters and the fidelity of fourteen different objects within a single workflow. This level of control is essential for those who need to build visual narratives or intricate storyboards, ensuring that element identities stay constant without involuntary alterations.
In addition to visual coherence, the model has made giant strides in text rendering, enabling perfectly legible and precise text to be inserted, with the added ability to translate and localize the text directly within the image.
From a technical standpoint, Nano Banana 2 offers production-ready specifications, ensuring full control over different aspect ratios and resolutions ranging from 512 pixels up to 4K. Visual content remains impeccably sharp whether used for vertical social media posts or as backgrounds for presentations on large screens.
The model distribution is already underway and involves Google’s entire ecosystem: from the Gemini app, where it replaces previous versions for the Fast and Thinking models, to Google Search, Ads and professional platforms like AI Studio and Vertex AI. Google has also strengthened its verification tools by pairing SynthID technology with C2PA credentials to give users a clear and contextualized view of how AI was employed.
