Adobe Firefly, the new AI from Adobe, is now capable of performing magic

If you’re an artist, graphic designer, or advertiser, you’ve probably used Illustrator on more than one occasion. Adobe‘s program allows users to create and modify vector graphics (SVG files), which means images that can be scaled as much as you want without losing resolution (a significant advantage compared to JPG or PNG when using images on various-sized media).

Adobe Illustrator DOWNLOAD

Well, if you thought Illustrator’s multiple capabilities were already impressive, hold on tight. Adobe has announced at their MAX conference that the program will now feature integration with a new language model from Firefly, the AI created by Adobe capable of generating all kinds of images, including vector images from now on.

Adobe describes the new model as “the world’s first generative AI model focused on producing vector graphics.” Similar to other Firefly language models, this new vector model has been trained using Adobe Stock’s extensive image database, Adobe’s large image library.

With just a simple text input (or prompt), the beta version of Firefly’s new vector model will allow us to create any vector image we desire in Illustrator. But here’s the thing – we’re not just talking about simple shapes (like logos), but compositions with various objects that can be individually modified, much like regular layers.

To develop this vector model in Firefly, Adobe reportedly used tens of millions of vector images from Adobe Stock to train the AI, as explained by Alexandru Costin, the company’s Vice President for generative AI and Sensei. Costin also noted that this process was quite a journey; the development team focused on creating a model that could generate vector images with the fewest possible anchor points.

Adobe Illustrator DOWNLOAD

Certainly, it’s a significant addition to a design program used by millions of people worldwide. If you wanted to create a vector image in Illustrator but found yourself in one of those creative slumps, now all you need to do is describe what you’re looking for, and then modify the image to your liking.