Google Launches Nano Banana Pro with Search-Grounded Image Generation

Google DeepMind unveiled Nano Banana Pro on November 20, 2025, a state-of-the-art image generation model built on Gemini 3 Pro that sets itself apart by integrating real-time Google Search data into visual creation. The model can produce studio-quality images up to 4K resolution while accurately rendering text in multiple languages, addressing one of generative AI's most persistent weaknesses.

The breakthrough feature is Google Search Grounding, which allows the model to pull verified, current information directly from the web before generating visuals. When users request data visualizations like "Show the top five billionaires by net worth today," the system fetches real-time financial data rather than relying on potentially outdated training information. This transforms the tool from a creative generator into what Google describes as a "mini data-visualization pipeline" for journalists, researchers, and educators.

Nano Banana Pro demonstrates impressive compositional capabilities, blending up to 14 input images while maintaining the consistency and resemblance of up to 5 people. The model includes professional-grade controls for lighting, focus, camera angles, and color grading, with support for multiple aspect ratios suitable for social media, print, and web applications. Users can transform scenes from day to night, adjust depth of field, or apply localized edits to specific image regions.

The rollout follows a tiered access strategy. Free users receive limited quotas before reverting to the original Nano Banana model, while Google AI Plus, Pro, and Ultra subscribers get expanded access. The model is now available in the Gemini app's "Create images" mode with the "Thinking" model selected, and has been integrated into Google Ads globally, Workspace tools like Slides and Vids, and developer platforms including Gemini API and Vertex AI. Google addresses transparency concerns through SynthID watermarking technology, embedding imperceptible digital signatures in all generated media. Users can now upload images to the Gemini app and ask whether they were created by Google AI. Free and Pro tier users see a visible Gemini sparkle watermark, while Ultra subscribers and developers receive unwatermarked images for professional use.

The model succeeds Nano Banana (Gemini 2.5 Flash Image), released earlier this year. While the predecessor excelled at quick edits and photo restoration, Nano Banana Pro targets professional designers, marketers, and content creators requiring high-fidelity visuals with accurate text rendering. Internal studies suggest prompts under 25 words achieve 30% higher compositional accuracy, indicating optimal usage patterns for maximum effectiveness.

Availability extends to creative professionals through Flow, Google's AI filmmaking tool, where Ultra subscribers gain enhanced precision over frame composition and scene design. Enterprise customers can begin building scaled solutions immediately in Vertex AI, with Gemini Enterprise access rolling out soon. The multilingual text capabilities enable marketing teams to generate localized content for international audiences without separate design workflows.