Views
No views yet
| Property | Value |
|---|---|
| Spatial compression | 8x |
| Temporal compression | 4x |
| Latent channels | 16 |
| Parameters | 346.6M (170.1M encoder, 176.5M decoder) |
1git clone https://github.com/Linum-AI/image-video-vae.git
2cd image-video-vae
3uv sync
4uv run python encode_decode.py --mode image --input examples/images/original/camel_closeup.jpg1uv run python encode_decode.py \
2 --mode image \
3 --input examples/images/original/camel_closeup.jpg
1uv run python encode_decode.py \
2 --mode video \
3 --input examples/videos/original/woman_in_breeze.mp4└── vae.safetensors # VAE model weights (1.3GB)1@online{image_video_vae_2026,
2 title = {VAE: Reconstruction vs. Generation},
3 author = {Linum AI},
4 year = {2026},
5 url = {https://www.linum.ai/field-notes/vae-reconstruction-vs-generation}
6}