Stable Video

Generates short videos from text or images using diffusion models. Stability AI’s first open generative AI video model based on the image model Stable Diffusion. A GPU with at least 8GB VRAM like an RTX 3080 works best for local runs though cloud APIs reduce hardware demands. Base clips max at 5 seconds but extensions and stitching in tools like ComfyUI allow building longer sequences. The open-source models are free via Hugging Face with optional paid APIs for easier scaling. It offers similar quality for shorts but adds open-source flexibility while Runway provides longer native clips via subscription. Videos export as MP4 files compatible with most editors supporting 24fps at 576x1024 resolution. Web demos and no-code interfaces like ComfyUI make it accessible though local setup involves some commands.