embg

vectorscan

A portable fork of the high-performance regular expression matching library

flash-attention

Fast and memory-efficient exact attention

CUDA-Programming

Sample codes for my CUDA programming book

ThunderKittens

Tile primitives for speedy kernels

triton-viz

kernl

Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.

mlir-tutorial

MLIR For Beginners tutorial

CuAssembler

An unofficial cuda assembler, for all generations of SASS, hopefully ：）

xformers

Hackable and optimized Transformers building blocks, supporting a composable construction.

Triton-Puzzles

Puzzles for learning Triton

optimized-routines

Optimized implementations of various library functions for ARM architecture processors

firecracker

Secure and fast microVMs for serverless computing.

composable_kernel

Composable Kernel: Performance Portable Programming Model for Machine Learning Tensor Operators

oss-fuzz-gen

LLM powered fuzzing via OSS-Fuzz.

libphonenumber

Google's common Java, C++ and JavaScript library for parsing, formatting, and validating international phone numbers.

thefuck

Magnificent app which corrects your previous console command.

pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

anki

Anki's shared backend and web components, and the Qt frontend

fb303

fb303 is a core set of thrift functions that provide a common mechanism for querying stats and other information from a service.

ConcurrencyFreaks

velox

A C++ vectorized database acceleration library aimed to optimizing query engines and data processing systems.

nccl

Optimized primitives for collective multi-GPU communication

StackRabbit

An AI for playing NES Tetris at a high level. Based primarily on search & heuristic, with high quality board evaluation through value iteration.

g6k

The General Sieve Kernel

lexical

Lexical is an extensible text editor framework that provides excellent reliability, accessibility and performance.

skybison

Instagram's experimental performance oriented greenfield implementation of Python.

llvm-tutor

A collection of out-of-tree LLVM passes for teaching and learning

lzbench

lzbench is an in-memory benchmark of open-source LZ77/LZSS/LZMA compressors

quickstack

A tool to take call stack traces with minimal overheads

skdb

SKDB is an embedded SQL database that stays in sync.