SPbLA is a sparse Boolean linear algebra primitives and operations for GPGPU computations. It comes as a stand-alone self-sufficient library with C API for high-performance computing with multiple backends for Nvidia Cuda, OpenCL and CPU-only platforms. The library has PyPI pyspbla package (Orachev et al., 2021) for work within a Python runtime. The primary library primitive is a sparse matrix of Boolean values. The library provides the most popular operations for matrix manipulation, such as

more »

... nstruction from values, transpose, sub-matrix extraction, matrix-to-vector reduce, matrix-matrix element-wise addition, multiplication and Kronecker product.