Google Levels Up Nano Banana. Meet Nano Banana Pro.

Google has pushed its image generation tools further with the release of Nano Banana Pro, an upgraded model built on Gemini 3 that promises sharper text rendering, richer compositions and more control for creators. The model is live in the Gemini app and is starting to arrive in AI Studio, Vertex and other Google products.

Google calls Nano Banana Pro its most advanced image system yet. It supports text rendering across multiple languages and can blend up to fourteen reference images into a single composition, depending on where it is used. The company says the model is designed to close the gap between raw ideas and professional-quality visual output.

To help users get there, Google has outlined seven prompting techniques. The guidance includes structuring prompts with clear elements such as subject, location, style and editing instructions, adding camera and lighting details, specifying aspect ratios and file formats, defining how text should appear and clarifying the purpose of each uploaded reference image.

The company also highlighted seven creative capabilities: producing images with crisp embedded text, using Gemini 3 Pro’s reasoning for more accurate scenes, translating and localising text inside visuals, applying studio-style lighting and camera control, resizing up to 4K with precise aspect ratios, blending six to fourteen images while maintaining character consistency and creating brand-safe assets for packaging and apparel.

Google is candid about current limitations. The model still struggles with very small text, intricate details, multilingual nuances, data-heavy diagrams and complex edits where artefacts can appear. Character consistency can also vary across multiple revisions. Google says improvements in these areas are already in progress.

Nano Banana Pro is powered by Gemini 3 Pro and is available now in the Gemini app, with wider rollout underway.