Overview
- Nano Banana Pro is available worldwide in the Gemini app by selecting the Thinking image model, with higher quotas for AI Plus, Pro and Ultra subscribers and slower generation tuned for professional use.
- Built on Gemini 3 Pro, the system aims for fact-aware outputs by interpreting real data and hand-written notes to create infographics and diagrams with readable, layout-accurate text and multilingual translation.
- The model can compose from up to 14 input images and keep up to five people consistent across outputs, with expanded controls for lighting, focus, perspective and style in 2K and 4K formats.
- Google is integrating the model into Ads, Workspace (Slides and Vids), NotebookLM, the Gemini API/AI Studio and the Flow storyboard tool, while Adobe has enabled it in Firefly and Photoshop.
- All generated images carry SynthID watermarks plus a visible Gemini star, and a reverse-image check is rolling out in Gemini, as independent tests report clear progress alongside errors in complex data visualizations.