Google has released its new generation image generation model, Nano Banana Pro, which is built on this week's announced Gemini 3. It supports up to 4K resolution, 14 object fusion, and maintains identity consistency for 5 people. It also adds online search and dual watermark verification with SynthID/C2PA. The pricing has been increased accordingly: $0.139 per image for 1080p/2K output, and $0.24 per image for 4K output, which is 3-6 times higher than the previous 1024px version at $0.039.
Key Features
- Professional Control: Users can specify camera position, depth of field, focus, lighting, and color grading in prompts. Default provides 6 high-fidelity images for selection.
- Online Generation: The model can search the web in real-time, automatically fetch recipes, news, or stock prices, and generate corresponding infographics. It has been integrated into NotebookLM, Flow, Slides, and Vids.
- Dual Watermark: SynthID invisible watermark + C2PA content credentials are both embedded. Users can upload images to verify whether they were generated or modified by Google's model in the Gemini App.
Subscription Strategy
- Free tier has limited usage, and users will be downgraded to Nano Banana after exceeding limits. Plus/Pro/Ultra tiers offer higher quotas, but specific upper limits have not been disclosed. However, Ultra can access 4K mode in Flow video tools and Antigravity IDE. Developers can access through Gemini API, Google AI Studio, and Workspace plugins, with charges based on the number of images and resolution.
