S

Stable Video Diffusion Img2vid

Developed by model-hub
Stable Video Diffusion (SVD) is a diffusion model capable of generating short video clips conditioned on static images.
Downloads 116
Release Time : 12/14/2023

Model Overview

This model is a latent diffusion model trained to generate short video clips (14 frames) from image conditions. It includes a fine-tuned f8 decoder to ensure temporal consistency.

Model Features

Image-conditioned generation
Generates coherent video clips conditioned on static images
Temporal consistency decoder
Uses a fine-tuned f8 decoder to ensure temporal coherence in generated videos
High-quality output
Demonstrates superior video generation quality compared to similar models in user studies

Model Capabilities

Generate short videos from static images
Maintain inter-frame coherence in videos
576x1024 resolution video generation

Use Cases

Research applications
Generative model research
Exploring the technical boundaries of video generation models
Model safety research
Studying safe deployment solutions for potentially harmful content generation
Creative applications
Artistic creation
Providing creative tools for artists and designers
Educational tools
Developing educational or creative tools
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase