Over the last few days, a mysterious AI tool called Nano-Banana went viral on social media. People shared strange and funny edits of themselves and their pets, amazed at how well the tool preserved faces while putting them into wild scenarios. Speculation ran wild about who was behind it.
Now the answer is out. Nano-Banana was actually owned by Google, and today it officially becomes part of the Gemini app. What looked like a quirky side project was in fact a preview of Google DeepMind’s new image editing model.
The upgrade focuses on consistency. Early Gemini editing sometimes struggled with likeness, but the new model makes sure people and pets look like themselves across multiple edits. That means if you give your chihuahua a tutu or try out a vintage haircut, the results stay believable.
Gemini’s new editing goes beyond novelty. You can blend multiple photos to create new scenes, build out a room step by step with multi-turn editing, or apply the style of one image to another object. For example, you could take the color of flower petals and apply it to rain boots or design a dress using the pattern of butterfly wings.
When you are finished, Gemini also lets you turn edited images into short videos. Every creation is stamped with both a visible watermark and Google’s invisible SynthID marker so it is clear the content was AI-generated.
So yes, the viral Nano-Banana tool was not a scrappy startup after all. It was Google’s test run for Gemini’s most advanced image editor, and now it is here for everyone to try.
