Google has launched Nano Banana 2, now officially known as Gemini 3.1 Flash Image, an updated AI image generation model available starting today. This new iteration builds upon the previous Nano Banana models, which were based on the 3.0 branch, and aims to deliver performance comparable to Nano Banana Pro but with the speed of the Flash variant. The model leverages the Gemini 3.1 Large Language Model to incorporate more advanced world knowledge from the internet, enabling it to render objects with enhanced fidelity and produce more accurate infographics. Furthermore, Google states that Nano Banana 2 achieves Pro-level text accuracy in its image outputs, addressing a common challenge in AI image generation.