Stability AI has released Stable Video Diffusion 4K, an open-source video generation model capable of producing ultra-high-resolution 4K video clips.
Model Details
- Native 4K output — 3840×2160 resolution video generation
- Open-source — released under the Stability AI Community License
- 8-second clips — generates up to 8 seconds of video at 24fps
- Frame consistency — improved temporal coherence across all frames
- Efficient inference — runs on a single A100 GPU
Capabilities
SVD 4K supports text-to-video, image-to-video, and video-to-video translation. It excels at landscape, architectural visualization, and cinematic nature scenes.
Community Impact
As an open-source model, SVD 4K enables researchers, indie developers, and artists to experiment with high-quality video generation without API costs.
Availability
Model weights available on Hugging Face. ComfyUI and Automatic1111 integrations available at launch.