Instructions to use nvidia/Cosmos3-Edge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Cosmos
How to use nvidia/Cosmos3-Edge with Cosmos:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
MLX ports for Apple Silicon (reasoner + video generator)
#64
by Paris99 - opened
Hi NVIDIA Cosmos team, thanks for Cosmos3-Edge! MLX ports:
Reasoner:
- netprojects/Cosmos3-Edge-mlx-8bit (3.1 GB)
- netprojects/Cosmos3-Edge-mlx-6bit (2.6 GB)
Video generator (Omni):
- netprojects/Cosmos3-Edge-Omni-mlx-8bit (5.0 GB)
- netprojects/Cosmos3-Edge-Omni-mlx-bf16 (7.6 GB)
On an M5 Max, the 8-bit reasoner matched NVIDIA's next token ~98.4% of the time, and a full image-to-video run took ~113 s.