Google has introduced a new AI tool named Whisk, designed to generate images using other images as prompts rather than relying solely on detailed text descriptions. With Whisk, users can provide images to specify the subject, scene, and style for their AI-generated artwork. The tool also allows you to prompt with multiple images for each of these three elements to create more tailored results. While text prompts can be added for extra detail, they are entirely optional.
If you don’t have images readily available, Whisk includes a dice icon feature that generates random images—likely AI-generated—for you to use as prompts. Additionally, a text box at the end of the process lets you include further specifications, giving users even greater control over the final image. Whisk’s innovative approach to image generation simplifies the creative process by prioritizing visual inspiration over lengthy text prompts.
Whisk generates images accompanied by corresponding text prompts. If you’re satisfied with the outcome, you can mark the image as a favorite or download it. To further refine the image, you have the option to input additional text or click on the image to edit the text prompt.
In a blog post, Google emphasizes that Whisk is intended for quick visual experimentation rather than precise, detailed editing. The company acknowledges that the tool might not always produce the desired results, which is why it provides options to modify the underlying prompts.
Whisk leverages the latest version of Google’s Imagen 3 image-generation model to blend three images: one for the subject, one for the scene, and one for the style. The model automatically generates a detailed caption based on the images you select, which helps guide Imagen 3 in remixing the photo. Additionally, Google unveiled Veo 2, the next iteration of their video generation model, which the firm claims can comprehend “the unique language of cinematography” and experiences hallucinations such as extra fingers “less frequently” than other models (OpenAI’s Sora is likely one of those other models).
Currently, Whisk is only available in the United States and can be accessed through the Google Labs site.
- Code Smarter, Not Harder: 5 Free AI Tools You Need to Know! - January 22, 2025
- Meta Takes on CapCut with Its New Video Editing App ‘Edits’ - January 20, 2025
- NovaSky Debuts Affordable Open-Source AI Model for Advanced Reasoning - January 14, 2025
