A collection of algorithms and data structures

BLAS-like Library Instantiation Software Framework

High-efficiency floating-point neural network inference operators for mobile, server, and Web

Acceleration package for neural networks on multi-core CPUs

Fast Clojure Matrix Library

Tuned OpenCL BLAS

Library for specialized dense and sparse matrix operations, and deep learning primitives.

BLISlab: A Sandbox for Optimizing GEMM

Multi-Threaded FP32 Matrix Multiplication on x86 CPUs

The HPC toolbox: fused matrix multiplication, convolution, data-parallel strided tensor primitives, OpenMP facilities, SIMD, JIT Assembler, CPU detection, state-of-the-art vectorized BLAS for floats and integers

A library and extension that provides objects for scientific computing in PHP.

Stretching GPU performance for GEMMs and tensor contractions.

💥 Fast matrix-multiplication as a self-contained Python library – no system dependencies!

Distributed Communication-Optimal Matrix-Matrix Multiplication Algorithm

Sparse matrix formats for linear algebra supporting scientific and machine learning applications

DBCSR: Distributed Block Compressed Sparse Row matrix library

Accelerated General (FP32) Matrix Multiplication from scratch in CUDA

Meta.Numerics is library for advanced numerical computing on the .NET platform. It offers an object-oriented API for statistical analysis, advanced functions, Fourier transforms, numerical integration and optimization, and matrix algebra.

[Experimental] LLVM-accelerated Generic Linear Algebra Subprograms