Google Nano Banana Pro: Advanced AI Image Generation with Gemini 3 Pro Architecture

Free and Pro users will see a small "sparkle" watermark on their images

Google has introduced Nano Banana Pro, their advanced AI image generation and editing tool built on the Gemini 3 Pro architecture. This launch follows several months after the original Nano Banana model debuted on Gemini 2.5.

While the previous Nano Banana could generate images, restore photographs, and create miniature figurines, the upgraded version performs these tasks with enhanced intelligence. Additionally, it can now produce infographics and diagrams, thanks to its improved comprehension of real-world information.

According to Google's blog post: "Nano Banana Pro can help you visualise any idea and design anything—from prototypes to representing data as infographics to turning handwritten notes into diagrams."

Features of Nano Banana Pro include combining up to 14 images into a single design and maintaining consistent appearances for up to 5 people across multiple images. It can transform sketches into products or convert blueprints into photorealistic 3D structures. Users can easily apply preferred visual styles, colors, and branding to their mockups.

Nano Banana Pro offers studio-quality editing controls, allowing you to select and modify specific areas without affecting the rest of the image. You can adjust camera angles, change focus, control lighting effects (such as transforming day into night), and add special effects like bokeh. It also provides color grading capabilities and allows exporting in 2K or 4K resolution, suitable for social media, websites, or print materials.

Google is implementing SynthID, an AI watermarking system, to ensure responsible use. Every created or edited image contains an invisible watermark verifying it was generated by Google AI. Free and Pro users will notice a small "sparkle" watermark on their images, while ultra subscribers have the option to remove this visible watermark.

A standout feature is Nano Banana Pro's integration with Google Search, enabling creation of images using real-time information like current weather or live sports scores. You can generate visuals for recipes or educational flashcards, effectively turning internet searches or facts into pictures and visual content.

A significant update is the model's ability to add text directly to images, ensuring clarity and readability regardless of length. Users can create posters, mockups, or designs with text in various fonts, styles, textures, and calligraphy. Thanks to multilingual capabilities, text can be generated in multiple languages.

Josh Woodward, Vice President of Google Labs and Gemini, told CNBC that "Nano Banana Pro's capabilities expand beyond its original iteration, which launched in late August."

To use Nano Banana Pro, open the Gemini app and select "Create images" using the "Thinking model." You can either type a description, upload an existing image, or blend up to 14 images into one composition. Customize by editing details, adding text, or adjusting camera angles. You can connect to Google Search to incorporate live information, then choose your resolution and aspect ratio before exporting the final image.

Free users receive a limited number of image creations, while paid subscribers get higher quotas and access to additional features.