Google continues to develop and expand its proposal of tools based on generative artificial intelligence and has presented Whiska new innovation from its experimental area Google Labs. This is a solution that allows you to generate images from other images, instead of prompts or text instructions.

As the company explains, Whisk is differentiated by a fast, creative and fun process, as it makes it easy to generate compositions by simply dragging other images. Allows you to include an image for the subjectanother for himto scene and another for him style. These are mixed and can be customized to create unique images.
However, text can also be used to define the details.

The system is based on Google’s artificial intelligence models. So, Gemini automatically writes a detailed title to images created in Whisk, which is then used to feed the company’s latest image generation model, Image 3. Google points out that the essence is captured, but not an exact replica, so that it is easier to continue generating images later.

Google warns that Whisk extracts only a few key features from the images uploaded by the user, so the results may differ from expectations. “We understand that these features may be crucial to your project and that Whisk may not be the one for you, which is why we allow you to view and edit the underlying prompts at any time.“, explains the technology.

In a statement, Google wanted to highlight the Whisk differentiation. He comments that in the first tests with artists and creatives, they have described the new development as a creative tool, not a traditional image editor. “We created it for quick visual exploration, not for perfect edits. It’s about exploring ideas in new and creative ways, allowing you to work with dozens of options and download the ones you like best”, comments the company.

At the moment, Whisk is only available through Google Labs in the United States.