OpenAI is making online fashion shopping more visual with a new virtual try-on feature inside ChatGPT. The tool lets users upload a selfie or full-body photo and see how selected clothes or accessories could look on them before making a purchase. The feature appears in shopping results through a new ‘Try On’ button, while users can also upload screenshots or images of clothing and ask ChatGPT to create a preview. The update arrives with a new Favourites option that allows shoppers to save products to their library. OpenAI says both features are rolling out globally. Follow along with the article to learn how to use it and how it works.
ChatGPT has recently introduced a new feature known as ‘Try On’ that can help the consumers in online clothing shopping be more visual. According to the company, the new Try On feature takes advantage of the ChatGPT Images 2.5 model, which was released in September. Furthermore, the company also clarifies that the model can produce more natural lighting and richer textures while following instructions more reliably and creating images faster.
If you’re looking to use the Try On feature, then you need to head to the ChatGPT platform and open a clothing product shown in ChatGPT shopping results. Now tap on the ‘Try On’ button. You’ll be required to upload a selfie or full-body photo of yours for reference.
ChatGPT will now generate the preview of you wearing that cloth. The speed of image generation will depend on the internet speed of the user.
Other than the Try On feature, the AI giant has also introduced a Favourites section that allows users to save products to their Library. Thanks to this, the saved products can be revisited later and can appear alongside generated try-on images. Do note that ChatGPT can also help users build an outfit from a style description or identify clothing items from photos of celebrity outfits.
Keep in mind that OpenAI has clearly stated that images uploaded from personal accounts can be used to improve its models unless users opt out. The feature has started rolling out globally and can be used instantly.