Google has unveiled a new version of its image generation model Nano Banana Pro, developed based on Gemini 3 Pro. The update opens up new possibilities for users, including photorealism, 4K support, artificial intelligence for watermarking, and generation based on real-time data.
This is reported by Business • Media
Enhanced Features and New Capabilities of Nano Banana Pro
Google has reported that Nano Banana Pro represents a significant advancement compared to previous versions. With deeper context processing, improved handling of text elements, and access to real-time information through integration with Google Search, the new model elevates visualization to a studio design level. Users can now create not only simple visual effects but also professional layouts, 3D visualizations, prototypes, and educational materials. The system employs “advanced reasoning” mechanisms and works with actual data, making images as relevant and contextual as possible.
Among the key features of the model are improved text rendering, deeper context, and a higher level of creative control. The technology ensures accurate and readable reproduction of text in various languages directly within the image. According to Google, Nano Banana Pro is optimized for creating images with precise textual content—from short slogans to lengthy paragraphs. The model excels in working with diverse fonts, textures, calligraphy, and content localization.
Integration with Real Data and Business Tools
The integration with the Gemini 3 Pro platform allows Nano Banana Pro to consider actual data by connecting real-time information from Google Search. This opens up opportunities for creating images that visualize weather conditions, sports results, recipes, as well as documentary and educational materials. The model is positioned as the best tool for generating context-rich infographics and explanatory materials based on reliable facts.
For editing and creative work, Nano Banana Pro offers studio tools: lighting adjustments (e.g., day-night, bokeh effect), angles, focus, color correction, local edits of specific parts of the image, and creation in 2K and 4K formats. The model is also optimized for working with group compositions—up to five individuals can be included in a frame while maintaining likeness, and up to 14 source images are available for blending.
The model’s rollout is happening in phases. Users gain access through the Gemini app in the “create image” tab with a “reasoning” mode. Free accounts have a limited number of generations, after which users transition to the standard version of Nano Banana. Higher quotas are available for Google AI Plus, Pro, and Ultra users. In the U.S., the model is available in AI Mode in search for Pro and Ultra users, and globally in NotebookLM for subscribers.
For businesses and developers, Nano Banana Pro integrates with Google Ads, Google Workspace (Slides and Vids), Gemini API, Google AI Studio, Vertex AI, Google Antigravity, and Flow for creating video content with Google AI Ultra.
“All media created with Google tools is embedded with an unobtrusive digital watermark SynthID.”
Additionally, in the Gemini app, users can check whether an image was created using Google AI. For most users, a Gemini sparkle watermark will be visible, except for Ultra users and developers.
In September 2025, the Nano Banana editor became a key driver of growth for Gemini downloads in the U.S. According to Appfigures, the number of installations increased by 45%—reaching 12.6 million—allowing the app to take first place in the App Store, surpassing ChatGPT.