Whisk: Google's New "Image-to-Image" AI Generation Tool 128

小编 · 2025-02-12

Google has launched a brand-new AI image generation tool called Whisk. This tool allows users to mix images of various styles without using prompts — just provide images. No traditional text prompts needed. Developed by Google Labs, Whisk combines Gemini and Imagen 3 models, capable of quickly combining styles from multiple photos into new stylized image outputs.

Whisk: Google's New "Image-to-Image" AI Generation Tool

Google has launched a brand-new AI image generation tool called Whisk, which allows users to blend images in various styles by simply providing images, without the need for prompts. There is no need for traditional text prompts. Developed by Google Labs, Whisk combines the Gemini and Imagen 3 models, enabling it to quickly merge the styles of multiple photos into a new stylized image output, generating new pictures.

For example, if you provide a photo of a person, a photo of a scene, and a style image, it can freely combine the person and the scene style together to produce a new scene-and-person photo in that style.

Whisk Experience URL:https://labs.google/fx/tools/whisk/unsupported-country

Whisk's Main Features and Characteristics:

Image-to-Image Functionality: The biggest highlight of Whisk is that it allows users to directly upload images to generate new images, without the need to write complex text prompts.
Multi-Image Upload and Style Fusion: Users can upload multiple images to define the subject, scene, and artistic style of the generated image. This multi-dimensional input method allows users to blend various elements together to create new image works.
Utilizes Advanced AI Models: Whisk leverages Google's Gemini and Imagen 3 models, which enhance the quality and speed of image generation, making the generated images more refined and creative.
Auto-Generated Text Prompts: After generating a new image, Whisk is also capable of automatically generating related text prompts.

Whisk also supports users to directly generate and quickly modify images during the creative process. This means users can not only upload images but also adjust the generated images by adding descriptive text, which adds flexibility and variety to the creative process.

In addition, Google clearly stated that Whisk is intended for rapid visual exploration rather than fine-grained editing. Therefore, the generated images may sometimes fall short of expectations, which is also one of the reasons why Google provides users with the ability to edit the underlying prompts to optimize the results.

Whisk is now available in more countries and can be used in over 100 countries. Previously, Whisk was only available for trial in the United States; users in other regions may have to wait.