Explore Projects
Discover 37 open source projects
Showing 1-20 of 37 projects
JuliaLang/julia
The Julia programming language for technical computing.
hpcaitech/ColossalAI
Colossal-AI optimizes large AI model training and inference with distributed computing and GPU acceleration.
volcano-sh/volcano
A Cloud Native Batch System for running AI/ML workloads on Kubernetes at scale.
NVIDIA/thrust
A C++ parallel algorithms library that provides high-performance GPU acceleration for scientific computing and data analysis.
spack/spack
A flexible package manager for building, installing, and managing software across multiple platforms.
ProjectPhysX/FluidX3D
A high-performance GPU-accelerated fluid dynamics simulation library for scientific computing and visualization.
arrayfire/arrayfire
ArrayFire is a general-purpose GPU library that provides a high-performance, cross-platform programming interface for developers.
flame/blis
BLIS is a high-performance, open-source library for linear algebra computations, targeting developers working on scientific and numerical computing applications.
apptainer/singularity
Singularity is a container engine for scientific and HPC workloads, focused on portability and reproducibility.
open-mpi/ompi
Open source MPI (Message Passing Interface) implementation for high-performance computing.
NVIDIA/libcudacxx
NVIDIA's C++ Standard Library for high-performance GPU computing on your system.
NVIDIA/cccl
NVIDIA's CUDA Core Compute Libraries for accelerated computing and GPU programming in C++
chapel-lang/chapel
Chapel is a high-performance, open-source programming language designed for productive parallel computing.
moosefs/moosefs
A scalable, fault-tolerant, and highly performing distributed file system for HPC and software-defined storage applications.
ashvardanian/less_slow.cpp
A collection of C++, CUDA, and assembly programming techniques for high-performance computing and low-latency systems.
RRZE-HPC/likwid
A performance monitoring and benchmarking suite for developers working with hardware and low-level systems.
NVIDIA/cub
NVIDIA's cooperative primitives for CUDA C++ to simplify GPU programming.
AdaptiveCpp/AdaptiveCpp
A community-driven C++ compiler for heterogeneous programming models like SYCL, C++ parallelism, and CUDA/HIP.
apptainer/apptainer
Apptainer is an open-source container system for Linux that enables reproducible, portable, and secure high-performance computing (HPC) environments.
DTolm/VkFFT
A high-performance Fast Fourier Transform (FFT) library for Vulkan, CUDA, HIP, OpenCL, Level Zero, and Metal.
Stay in the loop
Get weekly updates on trending AI coding tools and projects.