compile-time chosen simd bitwise ops with wide hardware support

noise_step: Training in 1.58b With No Gradient Memory

Demo project showing a single Rust codebase running on CPU and directly on GPUs

CH32V003 is an ultra-cheap RISC-V MCU with 2KB SRAM, 16KB flash, and up to 18 GPIOs that sells for under $0.10

MLIR+EqSat

IP addresses break, dial keys instead. Modular networking stack in Rust.

egg is a flexible, high-performance e-graph library

MSEP.one main application repository

Open Machine Learning Compiler Framework

Distributed Training Over-The-Internet

Safe, portable, high performance compute (GPGPU) kernels.

Multi-platform high-performance compute language extension for Rust.

Small scalar autograd engine, inspired from Karpathy's micrograd, with some additional features, such as more activation functions, optimizers and loss criterions.

Applied AI experiments and examples for PyTorch

Generate beautiful changelogs from your Git commit history

GPU code optimization using EVOlutionary computation

manipulate LLVM IR files

manipulate LLVM IR files

Actor runtime and distributed systems framework for Rust

Deep learning at the speed of light.

Kernel Tuner

Blue Engine is a general-purpose and easy-to-use graphics engine written in rust.

Simple examples for PTX code extraction from CUDA and OpenCL kernels.

Platform Security Assessment Framework

Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruction.

Easily create large video dataset from video urls

A massively parallel, high-level programming language

Foundation Architecture for (M)LLMs

Code repository for Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model.