Stable Video Diffusion is a narrower tool than most entries here: it only animates an existing still image, rather than generating from a text description. That makes it a common building block inside other pipelines rather than a standalone product.
Stable Video Diffusion
by Stability AI
Open-weight model that animates a single still image into a short clip — no text prompts.
- Image-conditioning only — cannot be controlled via text prompts
- Base model generates 14 frames; SVD-XT generates 25 frames, at 576×1024, 4 seconds or less
- Open weights under a non-commercial/research license; commercial use requires a separate Stability AI license
- Commonly run via Hugging Face, ComfyUI, or NVIDIA NIM