Google has introduced the Nano Banana Pro, an advanced image generation and editing tool powered by the Gemini 3 AI model. This tool is designed for professionals and aims to enhance the quality of image creation by providing features such as high-resolution outputs, text rendering on images, and the capability to merge multiple images into a cohesive design. According to Google representatives, it targets both professionals and those aspiring to elevate their design skills.
Using the Nano Banana Pro requires access through the Gemini app, where users can start the image creation process by inputting prompts. While the base version is free, users may encounter limitations unless they subscribe to Google AI Plus, Pro, or Ultra.
In testing its capabilities, a user uploaded a photo taken near the Brooklyn Bridge and requested a lighting change from day to night. The AI generated a convincing image that accurately depicted the change, handling details like the direction of cars effectively. In another exercise, the tool was asked to produce infographics detailing weather forecasts for specific cities. The output was visually satisfactory, with logical text and data, along with citations for validation.
However, the AI faced challenges with more complex requests. When tasked with summarizing a recent article on European AI regulations in a comic book format, the output did not align with the requested content, though subsequent attempts improved its accuracy.
The tool also showed divergence from initial inputs when creating greeting cards. An attempt to set a summer scene resulted in inaccuracies, such as unrealistic clothing rendering and errors in text. While the overall quality of images produced was commendable, definite limitations remained in certain tasks. This suggests that while Nano Banana Pro represents a significant upgrade, it still has room for refinement in various editing contexts.
Source: https://www.theverge.com/news/825667/google-nano-banana-pro-test