Adobe Firefly video model teased with AI-powered video generation feature

Adobe Firefly Video Model, an upcoming artificial intelligence (AI) model capable of video creation, was previewed by the company on Wednesday. The software giant first announced the under-development video model in April and has now shared more details about it. Large language models (LLMs) will be able to generate video from text prompts as well as image inputs. Users can also create videos from different camera angles, styles, and effects. The company also said that the video model will be available in beta later this year.

Adobe Firefly video model previewed

In a newsroom post, the company detailed the capabilities of the native AI video model. A YouTube video was also shared to showcase its features. Once launched, the Firefly Video Model will join Adobe’s existing generator models, including image models, vector models, and design models.

Based on the YouTube video, it appears that the Adobe Firefly video model can generate video from both text- and image-based input. This means users will be able to write a detailed prompt or share an image as a reference for the output video.

The company claimed that users will also be able to make complex requests like multiple camera angles, lighting conditions, style, zoom and speed. Notably, the AI-generated videos shared by the company appear to be on par with the videos teased with OpenAI’s Sora.

Additionally, the company also demonstrated the Generative Extend feature, which was first teased (but not demonstrated) in April. This feature essentially allows users to extend the duration of a shot by adding additional frames. These frames are generated using AI by taking references from the preceding and following frames. This can give editors the option to lengthen a video or allow the camera to linger on a shot for a few seconds longer.

Citing Alexandru Costin, VP of Generative AI at Adobe, The Verge reports that the maximum length of AI-generated videos is capped at five seconds, which is on par with similar tools available in the market. Notably, while the company said that the Firefly video model will be available as a standalone app, it will also be integrated within Creative Cloud, Experience Cloud, and Adobe Express Workflows.

Furthermore, the company claims that the AI ​​video model is “commercially safe” and has been trained only on licensed content, data available in the public domain, and data taken from Adobe Stock. The software giant also highlighted that the AI ​​models will not be trained on user data.