01Open source
Repos
Open-source repos worth knowing, by the problem they take on.
Science & Simulation
GEOS-Chem Science Codebase
GEOS-Chem is the science codebase for the GEOS-Chem atmospheric chemistry model, containing routines, run scripts, and interfaces for multiple wrappers. It is a major Earth system science codebase used in global chemistry simulations.
SAMRAI Framework
SAMRAI is a C++ framework for developing block-structured adaptive mesh refinement applications. It targets large-scale parallel simulations with coupled physics and is notable as LLNL's scalable HPC AMR infrastructure.
MuJoCo Batch Runner
mjbatch is a Python library for running thousands of MuJoCo physics simulations in parallel on CPU using a C++ thread pool. It is useful for reinforcement learning and model predictive control requiring massive parallel rollout.
PyGMTSAR InSAR
PyGMTSAR provides Python-based satellite interferometry (InSAR) processing for Sentinel-1 and NISAR data. It is notable for making InSAR accessible and scalable, with a successor framework using Zarr version 3 for cloud-ready workflows.
Open Data Cube
datacube-core is the Open Data Cube core library for analyzing continental-scale Earth observation data over time. It provides a geospatial processing framework for managing and accessing large satellite data archives.
AgentTorch Models
AgentTorch implements Large Population Models for differentiable simulation of millions of interacting autonomous agents. It uses autograd to enable learning in agent-based models, targeting collective dynamics like pandemics and climate change.
ESPResSo Package
ESPResSo is a molecular dynamics package for simulating coarse-grained bead-spring models in soft matter research. It is a long-standing research code used in physics, chemistry, and molecular biology.
OpenHPC Integration
OpenHPC provides pre-built components for deploying and managing HPC Linux clusters, including provisioning, resource management, runtimes, and scientific libraries. It is an essential packaging and integration repository for cluster infrastructure.
Ascent Visualization
Ascent is a lightweight in situ visualization and analysis runtime for multi-physics HPC simulations. Developed at LLNL, it enables many-core capable data processing without moving data off-node.
Avogadro Libraries
Avogadro libraries offer 3D rendering, visualization, analysis, and data processing for computational chemistry, molecular modeling, bioinformatics, and materials science. They are key, actively maintained libraries widely used in these fields.
PyOpenCL Integration
PyOpenCL provides Pythonic access to OpenCL for parallel computing on GPUs and CPUs, including arrays and algorithms. It is a foundational, actively maintained integration widely used in scientific computing.
VexCL Library
VexCL is a C++ vector expression template library for OpenCL, CUDA, and OpenMP that reduces boilerplate in GPGPU development. It supports multi-device and multi-platform computations and remains actively maintained.
GPU Array Library
ArrayFire is a general-purpose tensor library for parallel computing on CPUs, GPUs, and other accelerators, offering hundreds of functions in areas like linear algebra, machine learning, and signal processing. It is notable as an established GPU computing library for HPC and scientific workloads.
Python CUDA Integration
PyCUDA provides Pythonic access to NVIDIA CUDA with arrays and algorithms. It is notable as an essential CUDA integration for Python, widely used in GPU scientific computing.
R Data Transformation
collapse is a C/C++-based R package for fast data transformation and statistical computing with a class-agnostic architecture supporting many object types. It is notable for being highly optimized for scientific data analysis in R.
GPU Data Visualization
Datoviz is a Vulkan-based GPU visualization engine for interactive scientific data such as point clouds, images, meshes, and volumes. It is notable as a unique, high-performance, actively maintained tool for scientific rendering.
JavaScript Numerical Library
stdlib is a fundamental numerical library for JavaScript and TypeScript providing mathematical and statistical functions. It is notable for being a comprehensive, actively maintained library widely useful across scientific computing in JS/TS.
Universal Storage Engine
TileDB is an embeddable C++ storage engine for dense and sparse multi-dimensional arrays, dataframes, and key-value stores. It is notable as a universal array storage engine tailored for scientific and geospatial data.
Algebraic Multigrid Library
AMGCL is a header-only C++ library that solves large sparse linear systems using algebraic multigrid, usable as a standalone solver or preconditioner. It is notable as an exceptional, actively maintained black-box solver for PDE discretizations on unstructured grids.
OpenQuake Engine
OpenQuake Engine is open-source software for calculating and assessing seismic hazard, risk, and decision-making tools developed by the GEM Foundation. It is widely used, actively maintained, and substantial.
PySPH Framework
PySPH is a Python framework for Smoothed Particle Hydrodynamics simulations, with performance-critical parts in Cython and OpenCL. It is a real simulation engine that supports OpenMP, OpenCL, and MPI and is actively maintained.
FiPy PDE Solver
FiPy is a finite volume partial differential equation solver written in Python. It is a NIST-maintained research code that remains actively developed for solving PDEs.
Micromagnetic Simulator
MuMax3 is a GPU-accelerated micromagnetic simulator for magnetization dynamics. It is a domain-specific tool known for its performance and active development, with precompiled binaries and documentation available.
Multiphase Flow Solver
MFC is an exascale compressible multiphase flow solver in Fortran using Fypp metaprogramming. It ran the largest known public CFD simulation at 200 trillion grid points, scaling to over 43,000 GPUs and was a 2025 Gordon Bell Prize finalist.
Loopy Code Generator
Loopy is a transformation-based code generator for high-performance array-based CPU and GPU code. It is notable for enabling powerful compute optimizations and is actively maintained.
Gonum Numeric Libraries
Gonum is a set of numerical libraries for Go providing matrices, statistics, optimization, and more. It is a mature, widely used suite that is actively maintained with six-month releases.