xsimd
SIMD accelerator
Provides C++ wrappers for SIMD intrinsics and parallelized mathematical functions to accelerate code execution on various platforms.
C++ wrappers for SIMD intrinsics and parallelized, optimized mathematical functions (SSE, AVX, AVX512, NEON, SVE))
2k stars
72 watching
258 forks
Language: C++
last commit: 10 days ago
Linked from 1 awesome list
avxavx512c-plus-plus-11cppmathematical-functionsneonsimdsimd-instructionssimd-intrinsicsssesvevectorization
Related projects:
Repository | Description | Stars |
---|---|---|
jeffamstutz/tsimd | Provides C++ SIMD types and utilities for Intel CPUs to improve performance | 343 |
ermig1979/simd | An image processing and machine learning library optimized for high-performance computations on various x86/x64 and ARM architectures. | 2,067 |
p12tic/libsimdpp | A portable header-only C++ low level SIMD library providing a unified interface to various instruction sets. | 1,242 |
ashvardanian/simsimd | A high-performance library for optimized linear algebra and similarity calculations using SIMD instructions | 989 |
vcdevel/vc | A C++ library for explicitly data-parallel programming on modern CPUs and GPUs. | 1,456 |
simdutf/simdutf | Library providing fast Unicode functions and transcoding capabilities. | 1,172 |
scinim/vectorize | A backend for Nim that provides SIMD vectorization capabilities to improve performance of numerical computations. | 15 |
lemire/simdcompressionandintersection | A C++ library that enables fast compression and intersection of sorted lists using SIMD instructions. | 427 |
bytedance/sonic-cpp | A fast JSON serialization and deserialization library with SIMD acceleration | 855 |
microsoft/directxmath | A SIMD C++ linear algebra library for games and graphics apps | 1,551 |
vectorclass/version2 | A C++ library leveraging SIMD instructions to improve performance on modern CPUs. | 1,307 |
lemire/simdcomp | A library for fast integer compression using SIMD instructions | 490 |
piotte13/simd-visualiser | A tool to visualize and understand Single Instruction/Multiple Data (SIMD) code | 664 |
lemire/streamvbyte | Fast integer compression using SIMD instructions | 376 |
simd-everywhere/simde | A header-only library providing fast and portable implementations of SIMD instructions on various hardware architectures. | 2,400 |