Google Unveils Nano Banana Pro With Major Image-Generation Upgrades
A new update brings Nano Banana Pro, an upgraded image-generation model designed to offer sharper visuals, improved text rendering, and built-in web-search capabilities. Since it is powered by the recently released Gemini 3, it introduces features that create more detailed images, support more font and language styles, and allow users to generate text with higher accuracy. Additionally, the model can search online, so users can look up information, pull recipe steps, or even create flash cards from web results.
Nano Banana Pro focuses on professional-grade control, and users can now adjust camera angles, lighting, depth of field, focus, and color grading with greater precision. Moreover, it supports higher resolutions by moving beyond Nano Banana’s 1024 × 1024px limit and enabling 2K or 4K output. Although the new model provides far better image quality, it is slower and costlier. The original model cost $0.039 per 1024px image, while Nano Banana Pro costs $0.139 for each 1080p or 2K image and $0.24 for each 4K image.
Higher Fidelity, More Consistency, and Expanded Access
The model can now use six high-fidelity reference shots and blend as many as 14 objects in a single scene. Furthermore, it maintains appearance and resemblance accuracy for up to five people, which gives creators stronger consistency across projects. A demo app is available, allowing users to explore these capabilities firsthand.
Nano Banana Pro is now rolling out across multiple AI tools. The Gemini app will use it by default, though free-tier users can only generate a limited number of images before the system switches them back to the original Nano Banana. Meanwhile, subscribers to AI Plus, Pro, and Ultra will receive higher generation limits, and they will also access the model inside NotebookLM for additional workflows.
In search, the AI mode now supports Nano Banana Pro for Pro and Ultra subscribers within the U.S. Ultra subscribers can also use it in Flow, the video creation tool, and Workspace customers can access it inside Slides and Vids. Developers, meanwhile, can work with the model through the Gemini API, Google AI Studio, and the Antigravity IDE.
New Safety Tools and Broader Verification Support
The system now integrates SynthID into the Gemini app, giving users the ability to upload an image and check whether it has been created or modified by the company’s image models. As the feature evolves, support for C2PA content credential detection will be added to strengthen content verification. Because image authenticity concerns continue to rise, these additions aim to improve trust and transparency.
The story has been updated to reflect the approach to supporting C2PA.
