awesome-ai-cae
A curated list of 100+ AI-ready tools for Computer-Aided Engineering, ranked by an AI-Readiness Score (agent-callability: MCP, Python API, CLI, pip). CFD, FEA, SPH, DEM, differentiable simulation, neural operators, PINNs, MCP servers.
https://github.com/kimimgo/awesome-ai-cae
Last synced: 8 days ago
JSON representation
AI/ML for Simulation
Verified vs declared
- csml-rpi/Foam-Agent - AI agent for automated CFD workflows. LLM-driven OpenFOAM simulation setup and execution.
- deepmodeling/deepmd-kit - Deep learning for molecular dynamics. Neural network potentials for large-scale atomistic simulations.
- dynamicslab/pykoopman - Data-driven Koopman operator approximation. Dynamical system analysis and prediction from time series.
- dynamicslab/pysindy - Sparse Identification of Nonlinear Dynamics. Data-driven discovery of governing equations from measurements.
- google-deepmind/graphcast - Graph neural network for medium-range weather forecasting. Ten-day forecasts in under a minute (Nature 2023).
- google/jax-cfd - JAX-based CFD. Differentiable Navier-Stokes solvers. GPU-accelerated, auto-differentiable.
- Koopman-Laboratory/KoopmanLab - Koopman Neural Operator for mesh-free nonlinear PDE solving. Multi-scale decomposition.
- lululxvi/deepxde - Deep learning library for PDEs. PINNs, DeepONet. Backends: TensorFlow, PyTorch, JAX, PaddlePaddle.
- microsoft/aurora - Foundation model for Earth system prediction. Atmosphere, ocean, air quality. Pre-trained on ERA5 and CMIP6.
- microsoft/ClimaX - Foundation model for weather and climate. Pre-trained on CMIP6, fine-tunable for downstream tasks.
- NeuralOperator/neuraloperator - Neural operators in PyTorch. FNO, SFNO, UNO for learning PDE solution operators.
- NVIDIA/physicsnemo - Physics-ML framework (formerly Modulus). PINNs, neural operators, GNNs, diffusion models. Apache 2.0.
- Terry-cyx/MetaOpenFOAM - LLM-based multi-agent framework for CFD. Automated simulation pipeline from natural language.
- tum-pbs/PhiFlow - Differentiable PDE simulations. Fluid dynamics with TF/PyTorch/JAX. ML-physics hybrid workflows.
- microsoft/mattergen - Generative model for novel inorganic materials design. Diffusion-based crystal structure generation with target property conditioning.
- NVIDIA/earth2studio - AI-driven Earth system forecasting framework. Built-in model zoo (FourCastNet, Pangu-Weather, CorrDiff, GraphCast).
CAD & Geometry
Verified vs declared
- CadQuery/cadquery - Parametric 3D CAD scripting. Build models with Python, export STEP/STL/IGES. OpenCASCADE kernel.
- CadQuery/OCP - Python wrapper for OpenCASCADE via pybind11. Low-level foundation for CadQuery and build123d.
- FreeCAD/FreeCAD - Open-source parametric 3D CAD modeler. Part design, FEM workbench, BIM, path (CAM).
- gumyr/build123d - Modern Python CAD with algebraic geometry API. Successor to CadQuery with cleaner builder pattern.
- mikedh/trimesh - Load and manipulate triangular meshes. Boolean operations, ray tracing, convex hulls, format conversion.
- nschloe/pygmsh - Python interface for Gmsh. Scripted geometry + mesh generation with parametric control.
- Open-Cascade-SAS/OCCT - Open CASCADE Technology. Kernel for 3D surface and solid modeling, CAD data exchange (STEP/IGES).
- SolidCode/SolidPython - Python frontend for OpenSCAD. Generate 3D models programmatically with CSG operations.
CFD — Computational Fluid Dynamics
Verified vs declared
- OpenFOAM/OpenFOAM-dev - The open source CFD toolbox. Finite volume solvers for incompressible/compressible flow, multiphase, combustion, heat transfer.
- su2code/SU2 - Multiphysics simulation and design optimization. Compressible/incompressible flow, structural analysis, adjoint-based design.
- LLNL/Nek5000 - High-order spectral element CFD solver. DNS/LES of turbulent flows. Scalable to millions of cores.
- Nek5000/nekRS - GPU-accelerated spectral element CFD. Successor to Nek5000 with native CUDA/HIP/OpenCL support.
- precice/precice - Coupling library for multi-physics simulations. Fluid-structure interaction, conjugate heat transfer.
- PyFR/PyFR - High-order flux reconstruction CFD on mixed unstructured grids. GPU-accelerated (CUDA/OpenCL/HIP).
- ProjectPhysX/FluidX3D - GPU-accelerated Lattice Boltzmann fluid simulator. Real-time 3D visualization, scriptable via Python subprocess, supports multi-GPU.
Contributing
Verified vs declared
Data Formats & I/O
Verified vs declared
- nschloe/meshio - I/O for mesh formats. Abaqus, CGNS, Gmsh, VTK, XDMF, Exodus, and 30+ more.
- h5py/h5py - Pythonic interface to HDF5. Read/write large numerical datasets efficiently.
- Unidata/netcdf4-python - Python/NumPy interface to NetCDF. Climate, ocean, atmospheric simulation data.
- CGNS/CGNS - CFD General Notation System. Standard for CFD data storage and exchange. HDF5-based.
Datasets & Benchmarks
Verified vs declared
- divelab/AIRS - AI for science benchmarks. Molecular, protein, climate, physics datasets.
- Extrality/AirfRANS - RANS simulation dataset for airfoils. 1000 simulations with Reynolds-averaged fields (NeurIPS 2022).
- Mohamedelrefaie/DrivAerNet - Large-scale automotive CFD dataset. 4000+ car designs with drag coefficients and surface fields.
- i207M/PINNacle - Comprehensive PINN benchmark with 20 PDE problems across difficulty levels (NeurIPS 2024).
- NASA TMR - Turbulence Modeling Resource. Validation cases for CFD turbulence models with experimental data.
- pdebench/PDEBench - Benchmarks for scientific ML. Standardized PDE datasets with baseline models.
- PolymathicAI/the_well - Large-scale collection of diverse physics simulations for ML. Fifteen-plus PDE systems (NeurIPS 2024).
DEM — Discrete Element Method
Verified vs declared
- CFDEMproject/LIGGGHTS-PUBLIC - Industry-standard open-source DEM for granular materials. LAMMPS-based with heat transfer and CFD coupling.
- lammps/lammps - Large-scale Atomic/Molecular Massively Parallel Simulator. Classical MD and DEM with granular package. Sandia National Labs.
- MercuryDPM - Open-source DEM for granular and particle-laden flows. Coarse-graining, contact models, and the MercuryCG analysis toolkit.
- SudoDEM/SudoDEM - DEM for non-spherical particles. Polyhedra, super-ellipsoids, and cylinders for realistic granular simulations.
- Yade - Extensible open-source DEM framework. Python scripting, deformable particles, coupled DEM-FEM and DEM-fluid problems.
Differentiable Simulation
Verified vs declared
- Autodesk/XLB - Differentiable Lattice Boltzmann for physics-ML. Scales to billions of cells on multi-GPU.
- google/brax - Massively parallel rigidbody physics on accelerator hardware. Millions of steps/second on TPU.
- jax-md/jax-md - Differentiable, hardware-accelerated molecular dynamics. Runs on CPU/GPU/TPU via XLA.
- gbionics/jaxsim - Differentiable multibody dynamics engine. Hardware-accelerated robot learning and control via JAX.
- google-deepmind/mujoco - Multi-joint dynamics with contact. General-purpose physics engine for robotics, biomechanics, and control.
- NVIDIA/warp - Differentiable simulation and spatial computing. Reverse-mode AD, PyTorch/JAX interop.
- taichi-dev/taichi - Productive GPU programming with automatic differentiation. DiffTaichi for differentiable physics.
- tumaer/JAXFLUIDS - Fully-differentiable CFD solver for 3D compressible single-phase and two-phase flows.
- rtqichen/torchdiffeq - ODE and SDE solvers with automatic differentiation. Adjoint-based backpropagation through continuous-time dynamics.
FEA — Finite Element Analysis
Verified vs declared
- CalculiX - Free 3D structural FEM. Linear/nonlinear static, dynamic, thermal analysis. Abaqus INP compatible.
- dealii/dealii - Adaptive finite elements. Supports hp-refinement, multigrid, and parallel distributed computing.
- ElmerCSC/elmerfem - Multiphysics FEM solver. Fluid dynamics, structural mechanics, electromagnetics, heat transfer. CSC Finland.
- FEniCS/dolfinx - Next-generation FEniCS. Automated PDE solving with high-level Python/C++ interface. Parallel, scalable.
- firedrakeproject/firedrake - Automated FEM with code generation from high-level problem descriptions. UFL domain-specific language.
- FreeFem/FreeFem-sources - Partial differential equation solver using finite element method. High-level scripting language for 2D/3D problems.
- idaholab/moose - Multiphysics Object-Oriented Simulation Environment. Coupled physics FEM framework from Idaho National Lab.
- KratosMultiphysics/Kratos - Framework for multi-physics FEM. Structural, fluid, thermal, contact, FSI.
- mfem/mfem - High-order finite element library. Supports GPU acceleration, AMR, and dozens of physics applications.
- OpenSees/OpenSees - Open system for earthquake engineering simulation. Structural and geotechnical response analysis. Berkeley.
- sfepy/sfepy - Simple Finite Elements in Python. Solve PDEs by FEM in 1D, 2D, and 3D with plain Python scripting.
How the Score Works
Learning Resources
Verified vs declared
- barbagroup/CFDPython - Classic "12 Steps to Navier-Stokes" tutorial. Learn CFD fundamentals with Python step by step.
- ikespand/awesome-machine-learning-fluid-mechanics - Curated list of ML applications in fluid mechanics. Papers, code, and tutorials.
- jxx123/simglucose - Type 1 diabetes simulator. Example of AI-in-the-loop biomedical simulation.
- maziarraissi/PINNs - The foundational PINN reference implementation. Data-driven PDE solutions and discovery (JCP 2019).
- thunil/Physics-Based-Deep-Learning - Comprehensive collection of physics-based deep learning resources. Papers, code links, and tutorials from TUM.
- WillDreamer/Awesome-AI4CFD - Survey of ML for CFD covering data-driven surrogates, PINNs, and ML-assisted numerical solvers.
MCP Servers
Verified vs declared
- kimimgo/viznoir - Cinema-quality science visualization. 22 tools for rendering, slicing, contouring, volume rendering, and animating OpenFOAM/VTK/CGNS data via VTK. Headless EGL/OSMesa.
- llnl/paraview_mcp - Natural language control of ParaView via MCP. Multimodal LLM observes viewport for visual feedback (LLNL).
- webworn/openfoam-mcp-server - OpenFOAM MCP server with Socratic questioning for CFD education and expert error resolution.
Mesh Generation
Verified vs declared
- buaacyw/MeshAnything - Artist-quality mesh generation with autoregressive transformers. Any 3D input to mesh (ICLR 2025 spotlight).
- CGAL/cgal - Computational Geometry Algorithms Library. Mesh generation, triangulation, Boolean operations, convex hulls.
- Gmsh - Full-featured 3D finite element mesh generator. CAD engine, structured/unstructured meshing, built-in post-processing.
- libigl/libigl - Header-only geometry processing library. Mesh parameterization, deformation, Boolean ops. Eurographics award winner.
- MmgTools/mmg - Anisotropic mesh adaptation for 2D/3D surface and volume remeshing. Metric-based automatic refinement.
- NGSolve/netgen - Automatic 3D tetrahedral mesh generator. CAD (OCC) integration, mesh optimization, parallel meshing.
- nmwsharp/geometry-central - Applied geometry algorithms for surfaces and volumes. Geodesics, vector fields, intrinsic triangulations.
- OpenMeshLab/MeshXL - Foundation model for 3D mesh generation. Pre-trained on Objaverse, text-to-mesh capable (NeurIPS 2024).
- PyMesh/PyMesh - Geometry processing library. Boolean, convex hull, remeshing, self-intersection repair.
- pyvista/tetgen - Python interface to TetGen tetrahedral mesh generator. Constrained Delaunay tetrahedralization with quality control.
- wildmeshing/fTetWild - Fast and robust tetrahedral meshing. Handles self-intersections and degenerate input. Ten times faster than TetWild.
- buaacyw/MeshAnythingV2 - Adjacent Mesh Tokenization for efficient artist-quality mesh generation. Faster and higher-quality than V1 (ICCV 2025).
Optimization
Verified vs declared
- meta-pytorch/botorch - Bayesian optimization in PyTorch. Sequential decision making, multi-objective optimization, batch acquisition.
- OpenMDAO/OpenMDAO - Multidisciplinary design optimization. NASA-developed. Gradient-based + surrogate-assisted optimization.
- anyoptimization/pymoo - Multi-objective optimization. NSGA-II/III, reference directions, constraint handling, parallelization.
- dl4to/dl4to - Deep learning for 3D topology optimization. Autograd + adjoint method for efficient neural optimization.
- williamhunter/topy - Topology optimization with Python. Minimum compliance, heat conduction, mechanism design.
- mdolab/OpenAeroStruct - Aerostructural optimization. VLM aerodynamics + beam FEM structures + ply-level composites.
- OpenMDAO/dymos - Open-source optimal control of dynamic systems. Gradient-based trajectory optimization with Gauss-Lobatto and Radau collocation on OpenMDAO.
Quickstart — give your agent a CAE tool
SPH — Smoothed Particle Hydrodynamics
Verified vs declared
- DualSPHysics/DualSPHysics - GPU-accelerated SPH solver. Free-surface flows, wave generation, fluid-structure interaction, floating bodies.
- InteractiveComputerGraphics/SPlisHSPlasH - Physically-based SPH fluid simulation. DFSPH, IISPH, PBF pressure solvers. Viscosity, surface tension.
- pypr/pysph - SPH framework in Python. Compressible/incompressible flows, solid mechanics, coupled problems.
Surrogate Models & PINNs
Verified vs declared
- mathLab/PINA - Physics-Informed Neural networks for Advanced modeling. PyTorch Lightning-based with multi-device training.
- mathLab/PyDMD - Dynamic Mode Decomposition. Data-driven reduced-order modeling for fluid dynamics and beyond.
- NeuroDiffGym/neurodiffeq - Neural network solver for ODEs and PDEs. Flexible architecture with native boundary condition handling.
- NVIDIA/physicsnemo-sym - Symbolic AI for physics. Physics-informed neural networks with symbolic equation definition.
- rezaakb/pinns-torch - Production-ready PINNs in PyTorch. Multi-physics support, inverse problems, uncertainty quantification.
- sciann/sciann - Neural networks for scientific computing. Keras-based PINNs with custom loss and constraints.
- thuml/Neural-Solver-Library - Library for advanced neural PDE solvers. Benchmarking Transolver, FNO, and variants on diverse PDE families.
- camlab-ethz/poseidon - Scalable foundation model for PDEs. Pre-trained on diverse physics domains; few-shot generalization via in-context operator learning.
Visualization & Post-processing
Verified vs declared
- Kitware/VTK - The Visualization Toolkit. 3D computer graphics, image processing, scientific visualization. Industry standard.
- nmwsharp/polyscope - Lightweight 3D viewer for meshes, point clouds, and scalar fields. One-line visualization for geometry processing.
- pyvista/pyvista - Pythonic VTK. Streamlined 3D plotting, mesh analysis, and interactive visualization.
- Kitware/ParaView - Multi-platform data analysis and visualization. VTK-based GUI + Python scripting + client-server architecture.
- napari/napari - Fast n-dimensional image viewer. Plugin ecosystem for biomedical and scientific imaging.
- marcomusy/vedo - Scientific analysis and visualization of 3D objects and point clouds. VTK-based with simple API.
- plotly/plotly.py - Interactive, publication-quality graphs. 3D scatter, surface, mesh, volume. Web-based rendering.
- rerun-io/rerun - Multi-modal data logging and visualization SDK. Stream, store, and replay simulation data with Python API.
- fury-gl/fury - Free Unified Rendering in Python. VTK-based scientific visualization, 3D animations, and streamline rendering with a NumPy-friendly API.
- InsightSoftwareConsortium/itkwidgets - Interactive Jupyter widgets for 3D visualization of images, point sets, and meshes. Built on ITK and vtk.js.
- Kitware/trame - Build interactive scientific web applications purely in Python. Integrates VTK and ParaView for server-side or local 3D rendering.
- Kitware/VolView - Browser-based 3D radiological viewer for DICOM. Volume rendering, annotations, and measurements that run fully client-side.
- Kitware/vtk-js - Visualization Toolkit for the Web. WebGL/WebGPU scientific visualization and volume rendering entirely in the browser.