clSPARSE is an OpenCL™ library implementing sparse linear algebra routines.  It is the newest member of the OpenCL based clMathLibraries organization, and is a result of a collaboration between AMD Inc. and Vratis Ltd.  It exports a “C” interface allowing projects to build wrappers in any language they need.

clSPARSE features

  • Sparse Matrix – dense Vector multiply (SpM-dV)
  • Sparse Matrix – dense Matrix multiply (SpM-dM)
  • Sparse Matrix – Sparse Matrix multiply Sparse Matrix Multiply (SpM-SpM) – only single precision
  • Iterative conjugate gradient solver (CG)
  • Iterative biconjugate gradient stabilized solver (BiCGStab)
  • Dense to CSR conversions (& converse)
  • COO to CSR conversions (& converse)
  • Functions to read matrix market files in COO or CSR format

To learn more about clSPARSE, how to clone the code, download pre-built libraries or read the documentation, visit the project using our Github link in the upper right corner.


clSPARSE contains a directory of simple OpenCL samples that demonstrate the use of the API in both C and C++.  The cmake superbuild script builds the samples as an external project, in demonstration of how an application would find and link to clSPARSE with cmake.