Google Labs, the experimental arm of Google Testing a new renderer called Whisk. This tool allows people to recompose a photo by changing the subject, scene, and style, leading with images instead of text.

Whisk uses Google's rendering model, Imagen 3, to combine three images: one for the subject, one for the scene, and one for the style. For example, you can choose your own photo as the subject, a futuristic landscape as the scene, and an anime style for the final look.

The model automatically creates a detailed caption of your images, which is then used to guide Imagen 3 in creating a remix of the photo. You can also enter text prompts to further describe the desired outcome, including detailed descriptions such as "The subject rides a flying bike."

Because Whisk focuses on just a few key aspects of each image, the company explains that the results may not always meet your expectations. For example, the created object may have different height, weight, hairstyle or skin tone. Google says you can view and edit basic prompts at any time.

The trial is currently only available to users in the US at: labs.google/whisk.