We previously shared that Adobe’s generative AI tool for text-to-video would be coming to Firefly this year. Now, Adobe has finally released its long-awaited AI-powered video production tools to users. The company’s Firefly Video Model has launched with new tools integrated into Premiere Pro, offering users the ability to extend videos, generate videos from images, and create videos using text commands.

One of the standout features of Firefly is the Generative Extend tool, which can extend the beginning or end of videos by up to two seconds. This feature is particularly ideal for small adjustments and can be used to correct eye line shifts or unexpected movements. However, this extension feature is only applicable to short videos and works at resolutions of 720p or 1080p in 24 FPS format.

In addition to Premiere Pro, the Firefly web app also allows users to experiment with “Text to Video” and “Image to Video” tools. These tools generate short, low-resolution videos, such as five-second videos created from text commands, with a quality limited to 720p. Adobe indicates that these new tools are currently in beta and undergoing development.

Adobe has stated that its AI video tools are commercially safe to use. The company emphasizes that these tools have been trained on content that is not protected by copyright, presenting a more reliable option compared to other AI models.

This new AI technology was announced at the 2024 Adobe MAX conference, where Adobe invited users to adopt this technology. Alexandru Costin, Adobe’s Vice President of AI, stated, “We believe AI will accelerate the creative process,” adding that these new tools will greatly ease the content creation process for creative professionals.


Source