Nano Banana: A Visual Leap
Last year, Google leapfrogged in the AI race and much of the credit went to Nano Banana, its image generation model, which had become extraordinarily good. But its best features, like getting spellings correct in generated images or maintaining continuity of characters in multiple generations, were still reserved for the paid Pro version. Google has just released an update: Nano Banana 2, which now gives the prowess of the paid version free and is in fact better in many respects.
It is, for one, touted to be very fast. Another major plus point it has is advanced world knowledge, which means it can incorporate the latest information online into whatever image it is being asked to create, like infographics. As the company said in a blog announcing the release, “The model pulls from Gemini’s real-world knowledge base, and is powered by real-time information and images from web search to more accurately render specific subjects.” It can also do precision spelling and even translations of texts within images when editing them.
Google is attaching Nano Banana 2 to a number of its services. In its Gemini app, it will be the default. Those with paid subscriptions still get access to Nano Banana Pro for a regenerating option.