Stable Video Diffusion

video

by Stability AI · Updated June 1, 2025

Generate with Stable Video Diffusion

Stable Video Diffusion (SVD) is Stability AI's open-source video generation model that creates smooth, consistent video clips from a single reference image. It's the foundation for many community video generation tools and workflows.

Best For

Image-to-videoOpen-source workflowsCamera motionConsistent animationCommunity tools

Prompting Tips

  1. 1SVD is primarily image-to-video — start with a strong reference image
  2. 2Works best for camera motion and subtle scene animation
  3. 3Combine with ControlNet for more precise motion control
  4. 4SVD-XT generates 25 frames vs. standard 14 frames
  5. 5Use motion_bucket_id to control the amount of motion

Syntax & Constraints

Image-to-video model. Input: reference image. Parameters: motion_bucket_id (motion intensity), fps, num_frames (14 or 25 for XT). No text prompts — driven by the input image.

Build Prompts for Stable Video Diffusion

Other Stable Diffusion Models

Related Guides