Physics Informed Neural Networks Python Packages

pina-mathlab

Physics-Informed Neural networks for Advanced modeling

nops

nops is designed to be an elegant, high-level layer over complex neural operator architectures

pinnicle

A general PINN framework for solving ice sheet modeling

simulai-toolkit

A toolkit with data-driven pipelines for physics-informed machine learning.

neurodiffeq

A library for solving differential equations using neural networks based on PyTorch, used by multiple research groups around the world, including at Harvard IACS.

jwave

A JAX-based research framework for differentiable and parallelizable acoustic simulations, on CPU, GPUs and TPUs

pinnx

Physics-Informed Neural Networks for Scientific Machine Learning in JAX

tensordiffeq

Efficient and Scalable Physics-Informed Deep Learning and Scientific Machine Learning on top of Tensorflow for multi-worker distributed computing

pdebench

PDEBench: An Extensive Benchmark for Scientific Machine Learning

promethium-seismic

Promethium is a state-of-the-art, AI-driven framework for seismic signal reconstruction, denoising, and geophysical data enhancement, integrating cutting-edge deep learning architectures with production-grade data engineering.

parametrix

🪄 Flax-like computed parameters for bare JAX (and Equinox)

pinnstorch

An implementation of PINNs in PyTorch using Lightning and Hydra.

idrlnet

IDRLnet

fast-poisson-solver

A fast solver for the Poisson equation.

pararealml

A machine learning boosted parallel-in-time differential equation solver framework

oscillink

Oscillink — Self‑Optimizing Coherent Memory for Embedding Workflows

fastvpinns

A fast tensor-driven variational physics-informed neural network library for solving PDEs.

separable-operator-networks

Official repo for separable operator networks -- extreme-scale operator learning for parametric PDEs.

geoprior-v3

GeoPrior-v3: Physics-guided AI for geohazards

physdbd

Physics-based machine learning with dynamic Boltzmann distributions

pinnstf2

PINNs-TF2, Physics-informed Neural Networks (PINNs) implemented in TensorFlow V2.

pinnsjax

An implementation of PINNs in JAX.

frontx

⚛️ Nonlinear diffusion problems with JAX