Meet Whisk, Google's New Visual-First Approach to AI Image Generation

Meet Whisk, Google's New Visual-First Approach to AI Image Generation

Alongside the big unveiling of Veo 2 today, Google also rolled out an updated version of Imagen 3 globally, and introduced Whisk, a new playful image generation tool in Google Labs.

Key points:

  • Imagen 3 is rolling out globally to ImageFX users across 100+ countries with improved composition and style rendering
  • Whisk combines Imagen 3 with Gemini’s visual understanding to "remix" images
  • Users can mix and match subject, scene, and style images to create custom designs like digital plushies and enamel pins

The updated Imagen 3 model is now available through ImageFX in more than 100 countries. According to Google, the model produces brighter images with richer details and textures, while more accurately interpreting user prompts across a broader range of artistic styles – from photorealistic renders to impressionist artwork and anime. Here are some sample images:

Alongside this update, Google Labs is launching Whisk, a fresh take on AI image generation that moves away from traditional text prompts. Instead, you simply drag and drop reference images to define three key elements: the subject, the scene, and the style. This visual-first approach is funky and fun.

"We built it for rapid visual exploration, not pixel-perfect edits," explains Google in their announcement. The tool uses Gemini's visual understanding capabilities to automatically generate detailed captions of user-provided images, which then feed into Imagen 3 to create new variations.

Early testing with artists and creatives suggests Whisk fills a unique niche in the creative workflow. Rather than functioning as a traditional image editor, it serves as a rapid ideation tool, allowing users to quickly explore variations of concepts for products like digital plushies, enamel pins, and stickers.

Of course, since Whisk only extracts certain characteristics from reference images, generated results might differ from your expectations (which is half the fun). However, Google also provides you with the underlying prompt that Gemini generates so you can modify it and make refinements as desired.

Whisk is available exclusively to users in the US through Google Labs, where it joins the company's growing suite of experimental AI tools.

Chris McKay is the founder and chief editor of Maginative. His thought leadership in AI literacy and strategic AI adoption has been recognized by top academic institutions, media, and global brands.

Let’s stay in touch. Get the latest AI news from Maginative in your inbox.

Subscribe