Skip to content

setup.md

Latest commit

 

History

History
67 lines (45 loc) · 2.08 KB

setup.md

File metadata and controls

67 lines (45 loc) · 2.08 KB

Setup Guide

System Requirements

  • NVIDIA GPUs with Ampere architecture (RTX 30 Series, A100) or newer
  • NVIDIA driver >=570.124.06 compatible with CUDA 12.8.1
  • Linux x86-64
  • glibc>=2.31 (e.g Ubuntu >=22.04)
  • Python 3.10

Installation

Clone the repository:

git clone git@github.com:nvidia-cosmos/cosmos-transfer2.5.git
cd cosmos-transfer2.5

Install system dependencies:

uv

curl -LsSf https://astral.sh/uv/install.sh | sh
source $HOME/.local/bin/env

Install the package into a new environment:

uv sync
source .venv/bin/activate

Or, install the package into the active environment (e.g. conda):

uv sync --active --inexact

Downloading Checkpoints

  1. Get a Hugging Face Access Token with Read permission
  2. Install Hugging Face CLI: uv tool install -U "huggingface_hub[cli]"
  3. Login: hf auth login
  4. Accept the NVIDIA Open Model License Agreement.

Checkpoints are automatically downloaded during inference and post-training. To modify the checkpoint cache location, set the HF_HOME environment variable.

Advanced

Docker container

Please make sure you have access to Docker on your machine and the NVIDIA Container Toolkit is installed. To avoid running out of file descriptors when building the container, increase the limit with --ulimit nofile as in the example below.

Example build command:

docker build --ulimit nofile=131071:131071 -f Dockerfile . -t cosmos-transfer-2.5

Example run command:

docker run --gpus all --rm -v .:/workspace -v /workspace/.venv -it cosmos-transfer-2.5