Google has announced Nano Banana 2, officially called Gemini 3.1 Flash Image. This new model combines the speed of the earlier Flash version with the image quality of the Pro model.
Last year in August, Google introduced Nano Banana, also known as Gemini 2.5 Flash Image. It quickly became popular for image generation and editing. In December, Google launched Gemini 3 Pro Image, which offered better reasoning and more accurate results using Gemini’s advanced knowledge system.
Now, with Nano Banana 2, Google says users can get Pro-level image quality at Flash-level speed. The company also claims that this model matches state-of-the-art performance in image generation.
One of the biggest upgrades is support for real-world knowledge. Nano Banana 2 can use Gemini’s knowledge base, including real-time information from web search. This allows it to create more accurate visuals, including detailed infographics and data-based images.
Another major improvement is text generation. The model can now produce clear and readable text inside images. Text rendering has been a common problem for many AI image tools. Google says users can also translate and localize text within an image.
Nano Banana 2 also improves character and object consistency. It can maintain character resemblance for up to five characters in a single workflow. It can also preserve details for up to 14 objects at once. The model is better at following complex prompts more strictly.
The new version supports additional aspect ratios such as 4:1, 1:4, 8:1, and 1:8. It also supports image resolutions from 512 pixels up to 4K. According to Google, users will notice better lighting, richer textures, and sharper details compared to the previous Flash model.
Nano Banana 2 is now available for consumers, developers, and enterprises. In the Gemini app, it will replace the Pro image model as the default generator across Fast, Thinking, and Pro modes. It will also be available in Google Search AI Mode and Google Lens.
