Google Tests New Image Generator Whisk to Remix Three Images

Google Labs, the experimental arm of Google, is currently conducting tests on a novel image generator known as Whisk. This innovative tool presents a unique approach by enabling users to prompt with images rather than just text. It empowers them to remix a photo by making alterations to the subject, scene, and style.

Unlock the Potential of Image Remixing with Whisk

How Whisk Works

Whisk makes use of Google's image-generation model, Imagen 3. It combines three images - one for the subject, another for the scene, and one for the style. For instance, imagine selecting a photo of yourself as the subject, a captivating futuristic landscape as the scene, and an enchanting anime style for the final look. The model automatically generates a detailed caption for these images, which serves as a guide for Imagen 3 in creating a remarkable remix of the photo. Additionally, users can input text prompts to further define the desired outcome. They can provide detailed descriptions such as "Subject is riding a flying bike" to enhance the creative process.

Whisk's ability to focus on a few key characteristics from each image offers a new level of creative freedom. However, it's important to note that the results may not always meet one's expectations. For example, the generated subject might have differences in height, weight, hairstyle, or skin tone. But don't worry! Google allows users to view and edit the underlying prompts at any time, giving them the flexibility to fine-tune the remix according to their preferences.

The Current Availability

Currently, this exciting experiment is only accessible to users based in the United States. They can explore Whisk by visiting labs.google/whisk. It's an exclusive opportunity for American users to be at the forefront of image generation technology and experience the magic of remixing photos in a whole new way.

Google's foray into image generation with Whisk holds great promise. It showcases the company's continuous efforts in pushing the boundaries of technology and providing users with unique creative tools. As the experiment progresses, we can expect further enhancements and refinements to make Whisk an even more powerful and versatile image generator.