Adobe Firefly AI tool now lets you make videos from text

Adobe has expanded its generative AI toolkit to video as it bolsters its family of Firefly models with new updates.

Currently available in a limited beta, the Adobe Firefly Video Model is the first publicly available generative AI video model designed to be safe for commercial use. Firefly was already in place for creating images, with more than 13 billion images made since its first beta release in March 2023, but can now create video from text-based prompts as well.

Through the beta announced at Adobe MAX, Adobe is now seeking early feedback from a small group of creative professionals in the hopes of refining the model further. The beta test’s initial tool, Generative Extend is already available in Premiere Pro, allowing users to extend the end or beginning of footage for smoother transitions. It works best for minor adjustments but removes the need to reshoot footage when minor mistakes are made.

“The usage of Firefly within our creative applications has seen massive adoption, and it’s been inspiring to see how the creative community has used it to push the boundaries of what’s possible,” said Ely Greenfield, chief technology officer, digital media at Adobe, in a statement. “We’re thrilled to bring creative professionals even more tools for ideation and creation, all designed to be commercially safe.”

This push from Adobe is a sign of the company’s determination to stay competitive in an increasingly crowded market – something that will be especially compelling if future features launch in software like Premiere Pro for which users already have subscriptions.

Firefly can create video from both image and text-based prompts, as well as use text prompts to control the angle, motion, and zoom on videos. Users can write out a description of the video they need, and clarify the style, angle, and lighting, and Firefly can generate it just like with image generation. Images can also transform into virtually unlimited B-Roll or turn still shots into moving videos.

As well as generating fresh video, the Generative Extend tool helps to create perfectly timed video edits by extending clips to cover gaps in footage, smoothing out transitions, or holding shots longer as needed.

On top of launching these video tools, the latest Firefly updates also mean that image generation is four times faster than previously on the web app.

Featured image: Adobe

The post Adobe Firefly AI tool now lets you make videos from text appeared first on ReadWrite.