Skip to content

Stable Video Diffusion 1.1 Image-to-Video

Generate a short video from a single still image

Visit official site

What Stable Video Diffusion 1.1 Image-to-Video is

The model is described as a diffusion model that generates a video by taking a still image as a conditioning frame. It is a latent diffusion model trained to produce short video clips from an image.

What you can do with it

  • Generate short video clips from a still image
  • Apply the model to research on generative models
  • Explore and understand the limitations and biases of generative models
  • Safely deploy models that may generate harmful content

Model notes

  • The training setup targets 25 frames at 1024x576
  • Context frames of the same size are used in training
  • The model was fine-tuned from SVD Image-to-Video [25 frames]
  • During fine-tuning, 6FPS and Motion Bucket Id 127 were fixed to improve consistency without tuning hyperparameters

Example uses

  • Study the working principles and application areas of generative models
  • Use the model to produce design prototypes and artwork
  • Apply the model inside creative tools for education

Pros & cons

✓ What we like

  • Output consistency is improved without tuning hyperparameters
  • Short video clips are generated from a single image

! What to watch out for

  • Training is set to 25 frames at 1024x576
  • Fine-tuning fixes 6FPS and Motion Bucket Id 127 as conditions

FAQ

How many frames does the model generate?

The material states the model was trained to generate 25 frames at a resolution of 1024x576.

What was it fine-tuned from?

It was fine-tuned from SVD Image-to-Video [25 frames], with 6FPS and Motion Bucket Id 127 fixed during fine-tuning.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review