NVIDIA Containers

Help Desk

Updated: 12/01/2020

Nvidia delivers docker containers that contain their latest release of CUDA, tensorflow, pytorch, etc. You can see the full support matrix for all of their containers here:

Nvidia support matrix

Docker is not runnable on ALCF's ThetaGPU system for most users, but singularity is. To convert one of these images to singularity you can use the following command:


where $OUTPUT_NAME is typically of the form tf2_20.09-py3.simg and $NVIDIA_CONTAINER_LOCATION can be a docker url such as docker://nvcr.io/nvidia/tensorflow:20.09-tf2-py3

You can find the latest containers from Nvidia here: - Tensorflow 1 and 2 - Pytorch

For your convienience, we've converted these containers to singularity and are available here:


To extend the python libraries in these containers, please see building python packages.

For running with these containers, please see NVIDIA container notes.

For issues with these containers, please email support@alcf.anl.gov