Performance-portable, length-agnostic SIMD with runtime dispatch

Roaring bitmaps in C (and C++), with SIMD (AVX2, AVX-512 and NEON) optimizations: used by Apache Doris, ClickHouse, Alibaba Tair, Redpanda, YDB and StarRocks

Unicode routines (UTF8, UTF16, UTF32) and Base64: billions of characters per second using SSE2, AVX2, NEON, AVX-512, RISC-V Vector Extension, LoongArch64, POWER. Part of Node.js, WebKit/Safari, Ladybird, Chromium, Cloudflare Workers and Bun.

Portable wrapper for SIMD and vector instructions written in C++11. Compatible with NEON, SSE, AVX, AVX-512 and SVE (length specific).

Intel Homomorphic Encryption Acceleration Library accelerates modular arithmetic operations used in homomorphic encryption by leveraging AVX512 and IFM52 available on Intel's 3rd Generation Xeon Scalable Processors and later

A curated list of awesome SIMD frameworks, libraries and software

A general purpose machine code manipulation library for x86-32 (IA-32) and x86-64 (AMD64) architectures (Assembler, Disassembler, Library).

Masm compatible assembler

Fast C++ function "is_utf8": checks if the input is valid UTF-8. Made of a single source file. Optimized for ARM NEON, x64 SSE, AVX2 and AVX-512.

AVX-512 documentation beyond what Intel provides

(REOS) Radar and ElectroOptical Simulation Framework written in Fortran.

The fastest Run-Length-Encoding on the Planet (for x64)

Algorithms for matrix matrix multiplication, dgemm, AVX-256, AVX-512

DedupBench is a benchmarking tool for content-defined chunking techniques used in data deduplication. It currently supports eleven unique CDC techniques and five different vector instruction sets.

AVX-Hole C++ SIMD Library

A generic and efficient SIMD implementation of MSB Radix Sort with separate key and payload datastreams that supports arbitrary key and payload data types written in C++ accompanied by a bachelor's thesis.

Benchmark to show which is the fastest memcpy.

Runtime SIMD dispatch for Zig. Compile once per CPU level, pick the best at startup

Vector Dossier is a CLI tool that statically analyzes vectorization depth of programs and libraries

📚 Provides a handy way to hash data using Adler32, CRC (8-bit to 82-bit, and customizable without bit restriction), MD5, SHA-1, SHA-2 and SHA-3 (256-bit, 384-bit and 512-bit), including HMAC keyed hashing for some types. Functions for encrypting and decrypting data with Rijndael (128-bit, 192-bit and 256-bit) are also offered