Top 8 intrinsic Open-Source Projects

highway

66 3,645 9.8 C++

Performance-portable, length-agnostic SIMD with runtime dispatch

Project mention: Llamafile 0.7 Brings AVX-512 Support: 10x Faster Prompt Eval Times for AMD Zen 4 | news.ycombinator.com | 2024-03-31

The bf16 dot instruction replaces 6 instructions: https://github.com/google/highway/blob/master/hwy/ops/x86_12...

CTranslate2

13 2,776 9.0 C++

Fast inference engine for Transformer models

Project mention: Distil-Whisper: distilled version of Whisper that is 6 times faster, 49% smaller | news.ycombinator.com | 2023-10-31

Just a point of clarification - faster-whisper references it but ctranslate2[0] is what's really doing the magic here.
Ctranslate2 is a sleeper powerhouse project that enables a lot. They should be up front and center and get the credit they deserve.
[0] - https://github.com/OpenNMT/CTranslate2

InfluxDB

www.influxdata.com sponsored

Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
faster

1 1,548 0.0 Rust

SIMD for humans (by AdamNiederer)
simd_utils

1 80 6.6 C

A header only library implementing common mathematical functions using SIMD intrinsics
intel-intrinsics

1 66 8.6 D

The Dlang SIMD library

Project mention: SIMD with Zig | news.ycombinator.com | 2023-05-01

I wish more people understand that you absolutely need such intrinsics for fast software, there is no way around that.
https://github.com/AuburnSounds/intel-intrinsics

peakperf

2 56 4.4 C++

Achieve peak performance on x86 CPUs and NVIDIA GPUs
safe_arch

1 41 4.5 Rust

Exposes arch-specific intrinsics as safe function (via cfg).
WorkOS

workos.com sponsored

The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
camellia-simd-aesni

1 13 1.2 C

Camellia cipher SIMD vector implementations for x86 (with AES-NI, VAES and/or GFNI instructions), ARM (with ARMv8 Crypto Extension instructions) and POWER (with VMX+VSX+crypto instructions)

Project mention: Linux 6.5 Last Minute Fixes a Performance Regression, 34% Drop in a Benchmark | news.ycombinator.com | 2023-08-28

> camellia_aesni_avx_x86_64
An interesting point here is that AES-NI can be used to accelerate a host of things other than AES. In this case, it's because the S-box can take advantage of the AES S-Box (SubBytes) instruction: https://github.com/jkivilin/camellia-simd-aesni; https://kernel.googlesource.com/pub/scm/linux/kernel/git/sha....
Similar acceleration has been done with SM4, the Chinese analogue of AES. https://github.com/mjosaarinen/sm4ni

NOTE: The open source projects on this list are ordered by number of github stars. The number of mentions indicates repo mentiontions in the last 12 Months or since we started tracking (Dec 2020).