Google introduced a new virtual try-on capability for U.S. shoppers that allows users to generate a full-body digital version of themselves using only a selfie. The feature uses Nano Banana, Google’s Gemini 2.5 Flash Image model, to create studio-like images that can be used to preview clothing across Google Shopping. Users upload a selfie, select their typical clothing size, and choose from several generated images to set a default try-on profile.
The update expands Google’s existing virtual try-on system, which previously required a full-body photo or allowed shoppers to preview outfits on preset models with varied body types. By removing the need for a full-body image, Google aims to streamline the experience and make the tool more accessible for mobile shoppers.
The company said the feature supports billions of apparel listings connected to its Shopping Graph, offering a more personalized way to browse online clothing. Users can still opt to try on outfits using a full-body photo if they prefer. The announcement arrives the same day Google unveiled a more powerful version of its Gemini Deep Research agent, giving developers advanced autonomous research tools through the Interactions API and introducing the DeepSearchQA benchmark for evaluating multi-step web research agents.
