© Image generated by DALL-E AI for Presse-citron
AI image generators are not new: DALL-E 3 integrated into ChatGPT, Imagen 3, recently implemented in Gemini, Midjourney or Stable Diffusion. All of them use text to create more or less simple images. less successful: you type your query, and the image appears a few seconds later.
For its new tool, Whisk, Google has chosen to adopt a radically different approach by freeing itself from these textual constraints. More intuitive to use, it uses a universal language: that of the image. Explanations.
A creative architecture that is triply innovative
The singularity of Whisk lies in its tripartite methodology. The tool breaks down generation into three distinct dimensions: subject, scene, and style, each of which can be fed by multiple reference images. If you don't have an image in mind, Whisk's interface can generate one for you, and in a few clicks it will suggest illustrations (made by AI, of course) tailored to your request.
Powered by the latest version of the Imagen 3 model, Whisk simultaneously generates visuals and their associated text descriptions. Google emphasizes that the tool is designed for “rapid visual exploration, not pixel-perfect editing“. Generation times, although perceived as annoying by The Verge tester Jay Peters, do not seem prohibitive.
200% Deposit Bonus up to €3,000 180% First Deposit Bonus up to $20,000Faced with a result that does not exactly match expectations, Whisk allows you to gradually refine the generated image. It is possible to select a generated image, modify its underlying text prompt, or adjust reference images to guide the system to the desired result. This rapid feedback loop—a few seconds per generation—facilitates creative exploration through trial and error. As Google points out in its blog: “Whisk can sometimes miss its target ,” which is precisely why prompt editing is still available.
Whisk's intuitive interface allows users to shape unique creations by combining subject, scene, and visual style.© Jay Peters/the Verge
Alongside Whisk, Google announced that its Veo 2 model, capable of generating photorealistic videos, is coming in a new version. The latter would be better able to understand the “unique language of cinematography ” and would significantly reduce common and disturbing visual artifacts such as finger multiplication and other oddities, a recurring problem with competing models. This new evolution of Veo 2 will initially be deployed in VideoFX, accessible via Google Labs waiting list, before enriching YouTube Shorts ” and other products ” during 2025.
For the moment, neither Whisk nor Veo 2 are available in France or in Europe. The official Whisk website will greet you with this message: “Whisk is not yet available in your country“. After a few tries, even using a VPN didn't change anything and Google hasn't provided any official launch date for France.
- Whisk uses images as references to create new ones, without using text.
- The tool works in three steps: subject, scene and style, which can be modified at each iteration.
- Whisk and Veo 2 are not yet available in Europe.
📍 To not miss any Presse-citron news, follow us on Google News and WhatsApp.
M162.6 reviews
[ ]