Google introduced Nano Banana 2 based on Gemini 3.1 Flash Image with 4K support

Google introduced a new image generation model, Nano Banana 2, which operates on the Gemini 3.1 Flash Image platform. It is positioned as a faster version of Nano Banana Pro and will gradually replace it in the company's products.

What is known

According to Google, Nano Banana 2 retains the capabilities of the Pro model, including access to up-to-date information from web search for creating infographics, charts, and other visualizations, but performs tasks significantly faster. This allows for image editing and creating multiple variations with a single request.

The model also supports text generation on images for marketing materials, leaflets, or presentations. It is capable of maintaining character likeness for up to five characters in one workflow, which can be useful for storyboards and visual stories. Furthermore, Nano Banana 2 can execute complex instructions and generate images with resolution up to 4K with enhanced detailing.

Previously, Nano Banana Pro created such realistic images that they were difficult to distinguish from real ones. Due to high demand, Google even restricted its usage. Whether the new model will surpass the Pro in quality is still unknown.

Nano Banana 2 will become the default model in the Gemini app, AI Mode in Search, Google Lens, and the Flow AI creative studio. Meanwhile, subscribers of Google AI Pro and Ultra will retain access to Nano Banana Pro for specialized tasks.

Source: Google