Downloads · 30 days
0
0% of all-time downloads
cosmos-video2world
YuhDuh/cosmos-video2world
cosmos-video2world is a machine learning model from YuhDuh. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
After downloading, run bash reassemble.sh to combine the model parts into the full checkpoint.
All-time downloads
68
Public
Repo size
85.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.part_aa42.9 GB · 50%
At a glance
- Access
- Public
- Created
- Sep 3, 2025
- Updated
- Sep 3, 2025
- SHA
- 243aaaff
Model card
From the Hugging Face model README
COSMOS Video2World Model (14B)
Important: Reassemble the model
After downloading, run bash reassemble.sh to combine the model parts into the full checkpoint.
Model Details
- Architecture: 14B parameters
- Training iterations: 400
- Full checkpoint size: 80GB
- Split into 2 parts for Hugging Face upload limits
Usage
- Download all files
- Run
bash reassemble.sh - The model will be in
checkpoints/model/iter_000000400.pt
- Created
- Sep 3, 2025
- Updated
- Sep 3, 2025