Google has unveiled a new version of its AI image generation model, Nano Banana Pro. The update includes editing features, higher resolution, precise text rendering, and internet search capabilities.
We just dropped Nano Banana Pro, built on Gemini 3. 🍌
With state-of-the-art text rendering, vast world knowledge and studio-quality creative controls, Gemini 3 Pro Image can create and edit more complex visuals, infographics and more. Here’s what’s under the hood. 🧵 pic.twitter.com/x8fTe6pcmO
— Google DeepMind (@GoogleDeepMind) November 20, 2025
The neural network is based on the flagship LLM Gemini 3, introduced in November. The company claims that the Pro version surpasses its predecessor, Nano Banana, by creating more detailed images and accurate text in various styles, fonts, and languages.
The model now supports web search, allowing users to request a recipe and receive cards with its description.
Google stated that Nano Banana Pro is designed to give professionals greater control over images. It allows management of various aspects such as camera angle, scene lighting, depth of field, focus, and color gradient correction.
Nano Banana supports images up to 1024×1024 pixels. The Pro version offers 2K and 4K resolutions.
Gemini 3 Pro Image gives you a suite of controls in a range of accessible tools:
💡 Lighting and camera: change angles, focus, colors and more.
📈 Aspect ratio: export in a range of sizes, from 1:1 to 9:16 with ease.
💎 Higher resolution: the model supports outputs in 1K, 2K, and… pic.twitter.com/2zMAT0Nh03— Google DeepMind (@GoogleDeepMind) November 20, 2025
Drawbacks include a higher price and slower performance. The previous version charged $0.039 per 1024px image, while the new one requires $0.139 for 1080p or 2K and $0.24 for 4K.
Nano Banana Pro can place up to 14 objects in a single image while maintaining consistency. The tool is being integrated into many existing Google products.
Where to Try Nano Banana Pro
Gemini now uses the new model for image generation by default, but users without a paid subscription face limits on the number of images. Once the limit is reached, the system automatically switches to the Nano Banana model.
Subscribers to Google AI Plus, Pro, and Ultra will receive a higher limit and access to the model in NotebookLM.
Nano Banana Pro will be available in search via AI Mode for Pro and Ultra users in the U.S. Owners of the latter plan will be able to use the model’s capabilities in the company’s video service, Flow.
Developers can access the new tool through the Gemini API, Google AI Studio, and Antigravity.
The company is also implementing SynthID watermarking technology in the Gemini app. Users can upload an image to the chatbot, which will inform them if AI was used in its creation or editing.
Support for C2PA for content authenticity verification will be available in the future.
Back in September, the popularity of Nano Banana propelled it to the top of Gemini downloads.