Stability AI has officially introduced Stable Video Diffusion, a generative model for creating short videos. It is based on the Stable Diffusion model for creating images using text prompts. This is stated in a message on the company’s website.
Stable Video Diffusion is available only in the trial version and can create videos from a single image. The tool was released in two models – SVD and SVD-XT. They are respectively capable of creating 14 and 25 frames of video at a speed of 3 to 30 frames per second.
“Now available in research preview, this state-of-the-art generative AI video model represents a significant step in our journey toward creating models for everyone of every type,” Stability AI says in a statement.
The company also noted that at this stage, the model is not intended for real or commercial use. In addition, Stability AI is interested in user feedback on the safety and quality of their product to improve it for the final release.
Loading comments …