Overview

• Stability AI, the developer behind Stable Diffusion, has previewed a new generative AI that can create short-form videos with a text prompt, called Stable Video Diffusion.
• Stable Video Diffusion consists of two AI models, SVD and SVD-XT, capable of creating clips at a 576 x 1,024 pixel resolution, with frame rate speeds customizable between three and 30 FPS.
• The AI models have limitations, including not achieving perfect photorealism, inability to generate legible text, and difficulty with rendering faces.
• The project is currently in its early stages and is intended for research purposes only, with no plans for a wide release or for real-world or commercial applications at this time.
• Interested users can enter a waitlist for the preview of Stable Video Diffusion on the company's website.