PyG Release 24.05

NVIDIA Optimized Frameworks (Latest Release) Download PDF

This PyG container release is intended for use on the NVIDIA® Ampere Architecture GPU, NVIDIA A100, and the associated NVIDIA CUDA® 12 and NVIDIA cuDNN 8 libraries.

Driver Requirements

Release 24.05 is based on NVIDIA CUDA, which requires NVIDIA Driver release 535 or later. However, if you are running on a data center GPU (for example, T4 or any other data center GPU), you can use NVIDIA driver release 450.51 (or later R450), 470.57 (or later R470), 510.47 (or later R510), or 525.85 (or later R525), or 535.86 (or later R535). The CUDA driver's compatibility package only supports particular drivers. Thus, users should upgrade from all R418, R440, R460, and R520 drivers, which are not forward-compatible with CUDA 12.2. For a complete list of supported drivers, see the CUDA Application Compatibility topic. For more information, see CUDA Compatibility and Upgrades.

Contents of the PyG container

This container image includes the complete source of the NVIDIA version of PyG in /opt/pyg/pytorch_geometric. It is prebuilt and installed as a system Python module. The /workspace/examples folder is copied from /opt/pyg/pytorch_geometric/examples for users starting to run PyG. For example, to run the example:


/workspace/examples# python

See /workspace/ for details. The container also includes the following:

GPU Requirements

Release 24.05 supports CUDA compute capability 6.0 and later. This corresponds to GPUs in the NVIDIA Pascal, NVIDIA Volta™, NVIDIA Turing™, NVIDIA Ampere architecture, and NVIDIA Hopper™ architecture families. For a list of GPUs to which this compute capability corresponds, see CUDA GPUs. For additional support details, see Deep Learning Frameworks Support Matrix.

Key Features and Enhancements

  • GNN+LLM based Knowledge Graph RAG system calledG-retriever integrated into PyG. Example comes with a demonstration of GNNs reducing hallucinations of LLM. This enables a playground for users to combine any PyG GNN and Huggingface LLM.
  • GNN Platform removed


There are no announces for PyG in this release.

NVIDIA PyG Container Versions

The PyG container supports the same version of Ubuntu and CUDA as the PyTorch container.

Container Version Ubuntu CUDA Toolkit PyG PyTorch
24.05 22.04 NVIDIA CUDA 12.4.1 2.6.0 2.4.0a0+07cecf4
24.03 NVIDIA CUDA 2.5.0 2.3.0a0+40ec155e58
24.01 NVIDIA CUDA 12.3.2 2.4.0 2.2.0a0+81ea7a4
23.11 NVIDIA CUDA 12.3.0 2.4.0 23.11
23.01 20.04 NVIDIA CUDA 12.0.1 2.2.0 23.01

Known Issues

  • /workspace/examples/multi_gpu/ and can hit OOM on certain systems. If this issue is encountered NVIDIA recommends downgrading to the previous version of the container or upgrading to the latest, when available.

© Copyright 2024, NVIDIA. Last updated on May 29, 2024.