35 projects
Numba
Numba is a Just-In-Time (JIT) compiler that translates a subset of Python and NumPy code into fast machine code, specializing in numerical computing and scientific applications. It enables Python functions to be compiled to native machine instructions, significantly improving performance for computationally intensive operations.
Contributors
3,615
Organizations
871
Software value
$9.9M
CatBoost
CatBoost is a high-performance, open-source gradient boosting library developed by Yandex that implements gradient boosting on decision trees. It provides fast, scalable, and accurate machine learning algorithms for classification, regression, and ranking tasks, with built-in support for categorical features.
Contributors
3,537
Organizations
343
Software value
$242M
FlashAttention
FlashAttention is a high-performance implementation of attention for training deep learning models, offering faster and more memory-efficient attention computation compared to standard implementations. It achieves this through a novel algorithm that reduces memory access and increases hardware utilization.
Contributors
3,054
Organizations
487
Software value
$3.2M
Triton
Triton is a programming language and compiler framework designed for GPU programming, focusing on tensor computations and machine learning workloads. It enables developers to write high-performance code for GPU acceleration with Python-like syntax while providing low-level hardware control.
Contributors
2,866
Organizations
519
Software value
$7.3M
CUTLASS
CUDA Templates for Linear Algebra Subroutines
Contributors
1,341
Organizations
219
Software value
$60M
NVIDIA DALI
A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.
Contributors
1,259
Organizations
220
Software value
$14M
cuML
cuML - RAPIDS Machine Learning Library
Contributors
1,092
Organizations
166
Software value
$7M
CUDA.jl
CUDA.jl is a Julia programming language package that provides a comprehensive interface to NVIDIA's CUDA toolkit, enabling GPU computing capabilities within Julia. It allows developers to write high-performance GPU code using Julia's native syntax while abstracting away many low-level CUDA details.
Contributors
918
Organizations
233
Software value
$3.4M
SPIRV-Cross
SPIRV-Cross is a tool and library for performing reflection on SPIR-V and converting SPIR-V to other shader languages. It enables translation of SPIR-V shaders into GLSL, HLSL, MSL and other formats while preserving metadata and optimizations.
Contributors
773
Organizations
146
Software value
$10M
CUDA Core Compute Libraries (CCCL)
CUDA Core Compute Libraries
Contributors
766
Organizations
145
Software value
$27M
WarpX
The mission and charitable purposes (collectively, the “Charitable Purposes”) of the Project is to provide a community, performance-portable and modular Particle in-Cell code with advanced algorithms.
Contributors
599
Organizations
55
Software value
$12M
HOOMD-blue
Molecular dynamics and Monte Carlo soft matter simulation on GPUs.
Contributors
497
Organizations
37
Software value
$8.6M
QUDA
QUDA is a library for performing calculations in lattice QCD on GPUs.
Contributors
252
Organizations
28
Software value
$5.3M
Ginkgo
Numerical linear algebra software package
Contributors
208
Organizations
42
Software value
$9.7M
AMDGPU.jl
AMD GPU (ROCm) programming in Julia
This project hasn't been onboarded to LFX Insights.
Castro
Castro (Compressible Astrophysics): An adaptive mesh, astrophysical compressible (radiation-, magneto-) hydrodynamics simulation code for massively parallel CPU and GPU architectures.
This project hasn't been onboarded to LFX Insights.
CuPy
NumPy & SciPy for GPU
This project hasn't been onboarded to LFX Insights.
FlashInfer
FlashInfer: Kernel Library for LLM Serving
This project hasn't been onboarded to LFX Insights.
Futhark
:boom::computer::boom: A data-parallel functional programming language
This project hasn't been onboarded to LFX Insights.
GPUArrays
Reusable array functionality for Julia's various GPU backends.
This project hasn't been onboarded to LFX Insights.
HeAT
Distributed tensors and Machine Learning framework with GPU and MPI acceleration in Python
This project hasn't been onboarded to LFX Insights.
KernelAbstractions.jl
Heterogeneous programming in Julia
This project hasn't been onboarded to LFX Insights.
LinearSolve.jl
LinearSolve.jl: High-Performance Unified Interface for Linear Solvers in Julia. Easily switch between factorization and Krylov methods, add preconditioners, and all in one interface.
This project hasn't been onboarded to LFX Insights.
MIOpen
AMD's Machine Intelligence Library
This project hasn't been onboarded to LFX Insights.
Megatron-LM
Ongoing research training transformer models at scale
This project hasn't been onboarded to LFX Insights.
NCCL
Optimized primitives for collective multi-GPU communication
This project hasn't been onboarded to LFX Insights.
PIConGPU
Performance-Portable Particle-in-Cell Simulations for the Exascale Era :sparkles:
This project hasn't been onboarded to LFX Insights.
RAPIDS AI
Spark RAPIDS plugin - accelerate Apache Spark with GPUs
This project hasn't been onboarded to LFX Insights.
Torch-TensorRT
PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
This project hasn't been onboarded to LFX Insights.
VortexGPGPU-Vortex
VortexGPGPU-Vortex is a Linux Foundation project focused on developing a high-performance, open-source GPU architecture implementation that leverages General-Purpose GPU (GPGPU) computing capabilities for accelerating parallel processing tasks across various computing environments.
This project hasn't been onboarded to LFX Insights.
cuDF
cuDF - GPU DataFrame Library
This project hasn't been onboarded to LFX Insights.
nvFuser
A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")
This project hasn't been onboarded to LFX Insights.
rocBLAS
Next generation BLAS implementation for ROCm platform
This project hasn't been onboarded to LFX Insights.