Google Introduces Whisk AI Image Generation Tool

Google has introduced Whisk, a new tool allowing users to create images by combining elements such as subjects, scenes, and styles. This tool utilizes Google’s latest Imagen 3 model, which is capable of generating highly detailed and realistic images.

The Imagen 3 model combines Gemini’s visual understanding and description capabilities, allowing users to easily remix their ideas. Whisk lets users input or create images that convey the subject, scene, and style they have in mind, bringing them together to create something unique. The tool is part of Google Labs experiments and promises to enable users to prompt less and play more with their creativity. With Whisk, users can quickly visualize and remix ideas, making it a potentially powerful tool for artists, designers, and other creatives.

The release discusses two new tools from Google Labs: Imagen 3, an image generation model, and Whisk, a tool that allows users to prompt with images to visualize their ideas.

Imagen 3

• Achieved state-of-the-art results in side-by-side comparisons with leading image generation models
• Capable of generating highly detailed and realistic images with rich textures and compositions
• Examples of Imagen 3’s capabilities include:
  • A winter wonderland scene with a red squirrel holding a hazelnut
  • A craftsperson shaping pottery on a wheel with golden, luminous energy threads
  • A foggy 1940s European train station at dawn with intricate details and a melancholic atmosphere
  • A portrait of an Asian woman with neon green lights in the background
  • A close-up of a strawberry sculpted into the shape of a hummingbird in mid-flight
• Imagen 3 will be rolled out globally in ImageFX, Google’s image generation tool, to over 100 countries

Whisk

• A new experiment from Google Labs that lets users input or create images to convey their ideas
• Combines Imagen 3 with Gemini‘s visual understanding and description capabilities
• Allows users to remix subjects, scenes, and styles in fun and creative ways
• Users can drag and drop images, add written prompts, and refine their creations using a “refine” function
• The tool is designed to be playful and easy to use, with the slogan “prompt less, play more”

Both Imagen 3 and Whisk demonstrate significant advancements in image generation and manipulation capabilities, and have the potential to enable new forms of creative expression and exploration.