Adobe previews Firefly video model with text and image to video capabilities

Adobe Firefly

Adobe is bringing generative AI to videos with its next-generation Firefly Video model. The new model, part of Adobe’s Firefly family, introduces Text-to-video and Image-to-video capabilities, allowing users to generate videos from text prompts and transform still images into video clips.

Adobe has said that the new Firefly video model will be available to users by the end of the year.

Firefly video model: What all it can do
According to Adobe’s press note, the Firefly video model allows video creators to generate videos using text or change images to video and integrate them into existing workflow.

Users can generate videos with camera controls, such as angle, motion, and zoom, or reference images to create B-Roll footage. The model will also bring Generative Extend to Adobe Premiere Pro, providing video professionals with tools to enhance clips, fill gaps, and add new elements to footage.

These new capabilities will be available later this year on firefly.adobe.com and within Premiere Pro. Potential use cases include creating animations, adding video inserts with motion, and pairing AI-generated content with professionally captured footage to streamline storytelling.

Like other Firefly models, the Firefly Video Model is designed to be commercially safe, only trained on content Adobe has permission to use. Adobe emphasised that it does not use customer content for training its generative AI models.