Skip to main content
Ctrl+K
NVIDIA cuGraph - Home NVIDIA cuGraph - Home

NVIDIA cuGraph

  • Basics
  • nx-cugraph
  • Installation
  • Tutorials
  • Graph Support
    • WholeGraph
    • References
    • Developer Resources
    • API Reference
  • GitHub
NVIDIA cuGraph - Home NVIDIA cuGraph - Home

NVIDIA cuGraph

  • Basics
  • nx-cugraph
  • Installation
  • Tutorials
  • Graph Support
  • WholeGraph
  • References
  • Developer Resources
  • API Reference
  • GitHub

Section Navigation

  • Algorithms
    • List of Supported and Planned Algorithms
      • cuGraph Centrality Algorithms
        • Degree Centrality
        • Betweenness Centrality (BC)
        • Eigenvector Centrality
        • Katz Centrality
      • cuGraph Similarity Algorithms
        • Jaccard Similarity
        • Overlap Similarity
        • Sørensen Coefficient
      • cuGraph Community Algorithms
        • Louvain Community
        • Leiden Community
        • Spectral Clustering
        • Subgraph Extraction
        • Triangle Counting
        • K-Truss
  • Compatibility
  • Cugraph C++
    • cuGraph C++ Overview
  • Graph Neural Network Support
    • cugraph_pyg
    • WholeGraph
  • Data Stores
    • Knowledge Store
    • Feature Store
  • Graph Support
  • Graph Neural Network Support
  • WholeGraph

WholeGraph#

WholeGraph provides distributed memory, tensor, embedding, and graph operations for large-scale graph neural network workflows. It consists of:

  • pylibwholegraph, the Python and PyTorch-facing API; and

  • libwholegraph, the native C/CUDA library.

Both components are maintained in the cuGraph-GNN repository.

See the WholeGraph concepts, installation guide, and API reference.

previous

cugraph_pyg

next

Data Stores

NVIDIA NVIDIA
Privacy Policy | Your Privacy Choices | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2024-2026, NVIDIA Corporation.