Google releases Nano Banana Pro, its latest image-generation model

Google is upgrading its image-generation model with enhanced editing tools, higher resolutions, improved text accuracy, and even the ability to search the web.

Called Nano Banana Pro, the new model is built on Google’s latest large language model, Gemini 3, released earlier this week. Google says Nano Banana Pro offers more detailed visuals, better text rendering, and the ability to generate text in multiple styles, fonts, and languages — improving significantly on the original Nano Banana model.

Image Credits: Google

The updated model also supports web search, allowing users to perform tasks such as looking up a recipe and generating flashcards directly from the query.

Google says Nano Banana Pro is designed to give professionals more control over image creation, enabling adjustments to camera angles, scene lighting, depth of field, focus, and colour grading. And while the previous model was limited to 1024 × 1024px, Nano Banana Pro can generate 2K and 4K images.

Image Credits: Google

However, the improved quality comes at a higher cost and slower speed. The original model is priced at $0.039 per 1024px image, while Nano Banana Pro is priced at $0.139 for 1080p or 2K images and $0.24 per 4K image.

Nano Banana Pro can use up to six high-fidelity reference shots, blend up to 14 objects in a single image, and maintain visual consistency for up t5ve people. Google has also released a demo app showcasing these capabilities.

The new model is now rolling out across multiple Google AI tools. The Gemini app will use Nano Banana Pro as the default image generator. However, free-tier users can only access it for a limited number of creations before being switched back to the original model.
Google AI Plus, Pro, and Ultra subscribers will receive higher generation limits, though exact numbers haven’t been disclosed. These subscribers will also get access to Nano Banana Pro in NotebookLM.

In the U.S., AI Pro and Ultra subscribers can access the model through Search’s AI mode. Ultra subscribers will also be able to use it in Google’s video tool, Flow. The model is additionally available in Google Slides and Vids for Workspace customers.

Developers can use Nano Banana Pro through the Gemini API, Google AI Studio, and the company’s new IDE, Antigravity.

Google is also integrating SynthID, its watermarking and AI-image detection technology, directly into the Gemini app. Users can upload an image to check whether it was created or modified by Google’s image models.

Over time, Google plans to add support for C2PA content credential detection to strengthen content verification.

The story has been updated to reflect Google’s approach to supporting C2PA.