Google’s Whisk AI Experimental Tool Can Mash-Up Images to Generate Unique Outputs

Whisk uses Gemini and Imagen 3 models to create new images.

Written by Akash Dutta, Edited by Siddharth Suvarna | Updated: 18 December 2024 16:23 IST

Highlights

Google Whisk accepts both images and text as prompts
Users can add three images for subject, scene, and style
Once an image has been generated, users can refine and edit it

Google’s Whisk AI Experimental Tool Can Mash-Up Images to Generate Unique Outputs

Google’s Whisk tool is currently only available in the US via Google Labs

Photo Credit: Google

Google introduced a new experimental artificial intelligence (AI) tool on Monday that can fuse images to generate a unique output. Dubbed Whisk, it is a fun tool that does not have any larger application outside of its designated function. The Mountain View-based tech giant has released several such fun AI tools recently, such as GenChess, which uses the Imagen 3 AI model to generate unique chessboard pieces. With Whisk, the company is showcasing how AI can use just images as a prompt to generate unique art.

Google's Whisk Can ‘Remix' Input Images

In a blog post, the tech giant introduced the new AI tool. Whisk is currently only available in the US, and can be accessed via Google Labs, the company's platform to release experimental tools created using native AI models. Like all other tools, Whisk is also experimental and Google highlights that sometimes it may not perform the way users would like it to.

AI image generators are quite common, however, most of them either accept just text or a mix of text and images as input. In short, image generation models require natural language prompts in some capacity to understand what to create. However, Whisk is different from such models as users can add just images to prompt the model to create outputs.

Whisk asks users to add three images — one each for the subject, scene, and style. Once added, the AI tool automatically processes the visual information to generate a unique image which is the combination of all the three input images. Users can also add just two images, one for the subject and another for the scene, to generate output.

Google explained that behind the scenes, the Gemini model processes the images and writes a detailed natural language prompt, which is then fed to the Imagen 3 model. The prompt aims to capture the essence of the images and does not try to generate an objective blend of the input images.

Since Whisk is an experimental model, the generated images could be different from the user's expectations. To give users more control over the output, Whisk lets users refine and edit the images after generation. Users can easily check the underlying prompt written by Gemini and change it or add more information to get the desired result.

“We built it for rapid visual exploration, not pixel-perfect edits. It's about exploring ideas in new and creative ways, allowing you to work through dozens of options and download the ones you love,” Google said.

For the latest tech news and reviews, follow Gadgets 360 on X, Facebook, WhatsApp, Threads and Google News. For the latest videos on gadgets and tech, subscribe to our YouTube channel. If you want to know everything about top influencers, follow our in-house Who'sThat360 on Instagram and YouTube.