highway
brew install highway
v1.4.0
Apache-2.0 OR BSD-3-Clause
C++ header-only SIMD library providing portable vector intrinsics with runtime CPU dispatch across x86, ARM, WebAssembly, and other architectures.
Why you might care
Highway lets you write performance-critical code once and automatically dispatch to the fastest available CPU instructions (SSE4.2, AVX, AVX-512, NEON, etc.) at runtime. Unlike hand-rolled intrinsics or compiler autovectorization, it's predictable, maintainable, and works reliably across platforms. Essential for numeric libraries, video codecs, and data-parallel workloads that need 5–10x speedups with lower energy use.
Build dependencies
GitHub topics
Links
- https://github.com/google/highway
- GitHub: google/highway
- Brew formula source: Formula/h/highway.rb
Blurb generated by claude-haiku-4-5 on today.
Raw metadata
{
"aliases": [],
"alternatives": [
"Boost.SIMD",
"Vc",
"Eve",
"Intel IPP"
],
"build_dependencies": [
"cmake"
],
"categories": [
"library",
"header-only-library",
"simd"
],
"caveats": null,
"conflicts_with": [],
"dependencies": [],
"deprecated": 0,
"deprecation_reason": null,
"desc": "Performance-portable, length-agnostic SIMD with runtime dispatch",
"disable_reason": null,
"disabled": 0,
"enrichment_fetched_at": "2026-06-20T23:38:38+00:00",
"first_seen": "2026-06-20T23:34:18+00:00",
"full_name": "highway",
"github_default_branch": "master",
"github_last_commit_at": "2026-06-20T15:49:40Z",
"github_readme_excerpt": "# Efficient and performance-portable vector software\n\n[//]: # (placeholder, do not remove)\n\nHighway is a C++ library that provides portable SIMD/vector intrinsics.\n\n[Documentation](https://google.github.io/highway/en/master/)\n\nPreviously licensed under Apache 2, now dual-licensed as Apache 2 / BSD-3.\n\n## Why\n\nWe are passionate about high-performance software. We see major untapped\npotential in CPUs (servers, mobile, desktops). Highway is for engineers who want\nto reliably and economically push the boundaries of what is possible in\nsoftware.\n\n## How\n\nCPUs provide SIMD/vector instructions that apply the same operation to multiple\ndata items. This can reduce energy usage e.g. *fivefold* because fewer\ninstructions are executed. We also often see *5-10x* speedups.\n\nHighway makes SIMD/vector programming practical and workable according to these\nguiding principles:\n\n**Does what you expect**: Highway is a C++ library with carefully-chosen\nfunctions that map well to CPU instructions without extensive compiler\ntransformations. The resulting code is more predictable and robust to code\nchanges/compiler updates than autovectorization.\n\n**Works on widely-used platforms**: Highway supports seven architectures; the\nsame application code can target various instruction sets, including those with\n\u0027scalable\u0027 vectors (size unknown at compile time). Highway only requires C++17\n(language features, not necessarily the library) and supports four families of\ncompilers. If you want to use Highway on other platforms, please raise an issue.\n\n**Flexible to deploy**: Applications using Highway can run on heterogeneous\nclouds or client devices, choosing the best available instruction set at\nruntime. Alternatively, developers may choose to target a single instruction set\nwithout any runtime overhead. In both cases, the application code is the same\nexcept for swapping `HWY_STATIC_DISPATCH` with `HWY_DYNAMIC_DISPATCH` plus one\nline of code. See also @kfjahnke\u0027s\n[introduction to dispatching](https://g",
"github_repo": "google/highway",
"github_stars": 5635,
"github_topics": [
"avx",
"avx-512",
"avx-instructions",
"avx2",
"avx512",
"intrinsics",
"neon",
"simd",
"simd-instructions",
"simd-intrinsics",
"simd-library",
"simd-parallelism",
"simd-programming",
"sse42",
"wasm"
],
"homepage": "https://github.com/google/highway",
"homepage_og_description": null,
"homepage_og_image": null,
"homepage_title": null,
"installs_30d": 19853,
"installs_365d": 564817,
"installs_90d": 94284,
"keg_only": 0,
"keg_only_reason": null,
"last_seen": "2026-06-20T23:34:18+00:00",
"license": "Apache-2.0 OR BSD-3-Clause",
"llm_generated_at": "2026-06-20T23:43:37+00:00",
"llm_model": "claude-haiku-4-5",
"name": "highway",
"oldnames": [],
"one_liner": "C++ header-only SIMD library providing portable vector intrinsics with runtime CPU dispatch across x86, ARM, WebAssembly, and other architectures.",
"optional_dependencies": [],
"rank_30d": 236,
"rank_365d": 128,
"rank_90d": 188,
"raw_hash": "8d4ee18e3c485a1c",
"recommended_dependencies": [],
"revision": 0,
"ruby_source_path": "Formula/h/highway.rb",
"tap": "homebrew/core",
"test_dependencies": [],
"uses_from_macos": [],
"version_head": "HEAD",
"version_stable": "1.4.0",
"versioned_formulae": [],
"why_use_this": "Highway lets you write performance-critical code once and automatically dispatch to the fastest available CPU instructions (SSE4.2, AVX, AVX-512, NEON, etc.) at runtime. Unlike hand-rolled intrinsics or compiler autovectorization, it\u0027s predictable, maintainable, and works reliably across platforms. Essential for numeric libraries, video codecs, and data-parallel workloads that need 5\u201310x speedups with lower energy use."
}