Plug-in an image and it will attempt to give you a prompt to replicate that image, image to text model to get prompts from images.