Stable Video Diffusion: Image to Video Generation Model for ComfyUI

Stable Video Diffusion (SVD) is Stability AI's open image-to-video model, generating 14 or 25 frames from a single input image.

S

Stable Video Diffusion (SVD)

VideoImage-to-VideoOpen Source

Stability AI's image-to-video generation model. Takes an input image and generates a video by predicting frame sequences. Available in SVD (14 frames) and SVD-XT (25 frames) variants. Uses VideoLinearCFGGuidance for improved sampling quality.

Hugging Face (SVD)Hugging Face (SVD-XT)Research
DeveloperStability AI
Release Date2023-11 (SVD)
ArchitectureLatent Diffusion (UNet) — image-conditioned video
LicenseSVD License (see HF repo)
VariantsSVD (14 frames), SVD-XT (25 frames), SVD-XT 1.1
Key Parametersvideo_frames, motion_bucket_id, fps, augmentation_level

Guides and workflows related to this model series.

No articles found.