docker run --ipc=host --shm-size=512m --gpus all -it nvcr.io/nvidia/pytorch:24.02-py3 Use code with caution. Copied to clipboard
:In the extracted directory, run the following command to install Megatron-LM in editable mode: uv pip install -e . Use code with caution. Copied to clipboard Download Install Megatron Repo zip
For the most stable experience without manual dependency management, NVIDIA suggests using a pre-configured : docker run --ipc=host --shm-size=512m --gpus all -it nvcr
If you need additional features like , SentencePiece , or the Transformer Engine , use the following command for a complete setup: or the Transformer Engine