Google launches Nano Banana 2 with faster, high-quality image generation

Google has officially released the latest iteration of its image generation technology, dubbed Nano Banana 2, which boasts the capability to produce high-quality images at remarkable speeds. This upgraded model is fueled by the Gemini 3.1 Flash engine, which enhances realism compared to its predecessor. By integrating the nimble Gemini Flash AI model, the new image generator promises a more efficient and responsive experience, allowing for seamless back-and-forth iterations during the creation process. Notably, rather than merely relying on previously gathered training data, the model can now access real-time information from the internet. This means it can generate accurate images based on specific landmarks, individuals, or products. In addition to image generation, Google highlights that this advanced model can assist users in creating infographics, transforming notes into diagrams, and developing data visualizations. It also features capabilities for generating precise text, making it useful for tasks like marketing mockups or greeting cards, alongside functionalities for translation and localization. The new features of Nano Banana 2 have been detailed in a recent blog post from Google. Users can now maintain the appearance of up to five characters and incorporate as many as 14 objects within a single project. The model offers enhanced control over aspect ratios and resolutions, ranging from 512px to an impressive 4K. Moreover, the upgraded model enhances visual quality with improvements in lighting, richer textures, and sharper details—all while ensuring faster generation times. It is designed to handle both straightforward and intricate prompts, enabling users to request specific details and nuances with ease. Google emphasizes that Nano Banana 2 is engineered for rapid generation, accurate instruction adherence, and integrated image-search capabilities. The rollout of Nano Banana 2 is currently underway across various Google platforms, including the Gemini app, Search (in AI Mode and Lens), AI Studio, Gemini API, Google Cloud, Flow, and Google Ads.