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 📐

Library for specialized dense and sparse matrix operations, and deep learning primitives.

oneAPI Deep Neural Network Library (oneDNN)

Julia implementation for the BFloat16 number type

C++ template library for floating point operations

Half-precision floating point types f16 and bf16 for Rust.

A LLaMA2-7b chatbot with memory running on CPU, and optimized using smooth quantization, 4-bit quantization or Intel® Extension For PyTorch with bfloat16.

Customizable floating point types, with all standard floating point operations implemented from scratch.

Floating-Point Arithmetic Library for Z80

Comparison of PageRank algorithm using various datatypes.

Basic linear algebra routines implemented using the chop rounding function

Comparison of vector element sum using various data types.

IEEE 754-style floating-point converter

A JAX implementation of stochastic addition.

A Pytorch implementation of stochastic addition.

Round matrix elements to lower precision in MATLAB

Bfloat16 conversion utilities for Go/Golang

Hybridized On-Premise and Cloud (HOPC) Deployment Experimentation with Bfloat16