Adobe has confirmed that the Firefly Video Model will be released in a few months. Users will be able to try out a beta version before the end of the year. 

The Firefly Video Model will allow users to generate videos through text prompts. This video model was previously previewed earlier in the year. The model will expand Adobe’s suite of generative AI models. It will be able to support a broad multitude of cases including in the professional sphere or for more personal projects. 

Adobe has stated that the Firefly Video Model will come with three features namely Text-to-video, Image to Video and Generative Extend. All of these are in a private beta as of now but a public release is expected to roll out soon. Text-to-video allows video generation to take place with provided text prompts—similarly, image-to-video works with a similar premise. Videos generated will last around five seconds. Generative Extend is a feature that allows users to increase the total length of any provided video by two seconds. 

Firefly Video Model: More on this upcoming generative AI model

Firefly Video Model helps streamline the creative process and helps professionals in the fields envision their ideas. The video generation capability will be available through Premiere Pro, an Adobe video editing service. The model does not face many problems with the spelling of words, something prevalent in most generative AI models. 

This AI model is meant for commercial use so the model is trained on content that Adobe has received permission to use. Adobe has stated that the new video generation model does not train on content from Adobe customers. 

People interested in testing out the beta version can sign up for a waitlist on Adobe’s website. 

Suggested:

Generative Shape Fill: Adobe’s New AI Features.

Adobe Acrobat’s AI Assistant Summarises Documents & Meetings.