Up to 200x Faster Dot Products & Similarity Metrics — for Python, Rust, C, JS, and Swift, supporting f64, f32, f16 real & complex, i8, and bit vectors using SIMD for both AVX2, AVX-512, NEON, SVE, & SVE2 📐

Code for ACM MobiCom 2024 paper "FlexNN: Efficient and Adaptive DNN Inference on Memory-Constrained Edge Devices"

NEON ARMv8 SHA3_2x: 2 times SHA3 or SHAKE128/256 in 01 call. Use In Post-Quantum Cryptography Submission

Simple neural network microkernels in C accelerated with ARMv8.2-a Neon vector intrinsics.

Pipelined lowlevel implementation of COLM for ARM-based systems

CLAY - A C23 Linear Algebra librarY

arm neon experiment