Google Unveils Nano Banana Pro: 4K AI Images, Perfect Text, and Gemini 3 Core
Google has unveiled an enhanced version of its image-generation system, Nano Banana Pro — an evolution of last year’s Nano Banana, now offering far greater flexibility and markedly superior output quality. Built on the new Gemini 3 core, the model not only produces images but handles text with far greater precision, supports a wide variety of styles and fonts, and, when needed, can even search the web on its own. For example, you can ask it to find a recipe and instantly assemble illustrated cards based on the results.
The developers placed particular emphasis on controllability. Nano Banana Pro allows users to adjust camera angle, lighting, depth of field, object placement, and overall color palette — almost like working in a full-fledged editor. It also adds support for 2K and 4K resolutions, whereas the previous version was limited to 1024×1024 pixels. This additional clarity is invaluable for those creating presentations, instructional visuals, or content for large displays.
These expanded capabilities come at the cost of speed and price. The new model runs more slowly and is more expensive: instead of the previous 3 cents for a basic-format image, users must now pay $0.139 for 2K output and $0.24 for 4K. Google considers this a reasonable compromise given the added tools and heightened level of detail.
Nano Banana Pro can use up to six reference images to more accurately reproduce a character or preserve subtle facial features. A single frame can now include up to fourteen objects, and the system can maintain consistent likenesses for up to five people — a useful advantage for projects where characters must remain visually coherent from scene to scene. Google has opened a demo application where these functions can be tested immediately.
The new model is gradually being integrated into numerous Google services. The Gemini app now employs it by default, though free-tier users can generate only a limited number of images before being switched back to the basic Nano Banana. Subscribers to AI Plus, Pro, and Ultra receive higher limits, along with access to the model in NotebookLM.
In Google Search, Nano Banana Pro will appear in AI Mode for Pro and Ultra subscribers in the United States. Ultra users will also be able to employ it within the Flow video editor. Workspace customers will gain support in Slides and Vids, where the model will become a built-in component of presentations and video projects.
Developers already have access to Nano Banana Pro: the model is available through the Gemini API family, Google AI Studio, and the new Antigravity environment designed for those building applications atop generative systems.
Support Our Threat Intelligence
If you find our technology report and cybersecurity news helpful, consider supporting our work.
Buy Me a Coffee
PayPal
