Google has announced a new artificial intelligence tool called Whisk which allows you to generate images using other images as messages instead of requiring a long text message.
With Whisk, you can offer images to suggest what you'd like as the theme, scene, and style of your ai-generated image, and you can request multiple images from Whisk for each of those three things. (If you want, you can also fill in the text prompts.) If you don't have images on hand, you can click the dice icon to have Google fill in some images for directions (although those images will also appear to be generated by ai). You can also enter some text in a text box at the end of the process if you want to add additional details about the image you're looking for, but it's not required.
Whisk will then generate images and a text message for each image. You can bookmark the image or download it if you're happy with the results, or you can refine an image by entering more text in the text box or by clicking on the image and editing the text message.
In <a target="_blank" href="https://blog.google/technology/google-labs/whisk-visualize-and-remix-ideas-using-images-and-ai/”>a blog postGoogle emphasizes that Whisk is designed for “quick visual exploration, not pixel-perfect edits.” The company also says that Whisk can “miss the mark,” which is why it allows you to edit the underlying prompts.
In the few minutes I've used the tool while writing this story, it's been fun to play with. The images take a few seconds to generate, which is annoying, and although the images have been a little strange, it's been fun to repeat everything I've generated.
Google says Whisk uses the “latest” iteration of its image generation model Image 3, <a target="_blank" href="https://blog.google/technology/google-labs/video-image-generation-update-december-2024/”>what he announced today. Google also unveiled Veo 2, the next version of its video generation model, which the company says understands “the unique language of cinematography” and features things like extra fingers “less frequently” than other models (one of those other models is probably Sora from OpenAI). Veo 2 will come first to Google's VideoFX, for which you can get it on the Google Labs waitlist, and will expand to “other products” from YouTube Shorts sometime next year.
