cuTENSOR: A High-Performance CUDA Library For Tensor Primitives

Welcome to the cuTENSOR library documentation.

cuTENSOR is a high-performance CUDA library for tensor primitives.


Key Features

The documentation consists of three main components:

  • A User Guide that introduces important basics of cuTENSOR including details on notation and accuracy.

  • A Getting Started guide that steps through a simple tensor contraction example.

  • An API Reference that provides a comprehensive overview of all library routines, constants, and data types.


  • Supported SM Architectures : SM 6.0, SM 7.0, SM 8.0

  • Supported OSsRHEL 7/8, openSUSE 15, SLES 15, Ubuntu 20.04/18.04, Windows 10
    • Deprecated OSs : Ubuntu 16.04

  • Supported CPU Architectures : x86_64, ARM64, OpenPOWER


  • Dependencies : cudart, cutensor.h headers

Indices And Tables