Nano Banana Pro Debuts In Gemini 3 Pro Update

Google has introduced Nano Banana Pro, a new image-generation model released as part of the broader Gemini 3 Pro update. Announced on Thursday (20), the model focuses on producing images with greater detail and improving the accuracy and variety of text generated within images. This includes support for more diverse fonts, styles, and languages, enabling the system to create complex visuals such as diagrams and infographics with higher precision.

According to Google, Nano Banana Pro—also referred to as Gemini 3 Pro Image—can generate images that contain clearer and more accurate embedded text. This represents an important enhancement compared to previous versions, which sometimes struggled with typography consistency. As a result, the model can generate content that resembles structured design work, making it suitable for diagrams, educational material, and more professional visual outputs.

One of the notable features of Nano Banana Pro is its integration with Google Search. The model uses Search as a data foundation to draw relevant information for the requested image. This integration enables more contextually aligned visual outputs. Additionally, the system is capable of processing up to 14 separate input images simultaneously when creating an infographic or composite visual.

Nano Banana Pro also allows users to modify different elements of an image in ways similar to traditional editing software. Available adjustments include depth of field, focus, lighting effects, and color correction. These edits can be made while maintaining resolutions up to 4K, ensuring high-quality results without degradation.

To address concerns about misinformation and deepfake content, Google confirmed that all images generated with Nano Banana Pro include metadata compliant with the Content Provenance and Authenticity Coalition (C2PA) standard. This embedded data allows the detection of AI-generated images and helps users verify whether an uploaded picture has been digitally altered. In practice, users can upload an image to the Gemini app and ask whether it was generated by artificial intelligence.

The new model is now available for testing through the Gemini app. Users can access it by selecting the “Reasoning” model, which corresponds to Gemini 3 Pro, and switching to the “Create Images” mode. Free-tier users receive limited usage credits and will revert to the original Nano Banana model once those credits are depleted. Subscribers to Google AI Plus, Pro, or Ultra plans will have expanded usage allowances.

Features from the previous generation of the model are already integrated across multiple Google products, such as Search, Photos, and NotebookLM, continuing Google’s expansion of AI-powered tools across its ecosystem.

Filed in Robots. Read more about AI (Artificial Intelligence) and Google gemini.