PINA and PINN

These are competitors offering alternative PyTorch-based implementations of PINNs at different maturity levels—PINA provides a more comprehensive framework with advanced modeling capabilities, while the simpler implementation serves educational or minimal-dependency use cases.

PINA
57
Established
PINN
46
Emerging
Maintenance 10/25
Adoption 10/25
Maturity 16/25
Community 21/25
Maintenance 0/25
Adoption 10/25
Maturity 16/25
Community 20/25
Stars: 719
Forks: 95
Downloads:
Commits (30d): 0
Language: Python
License: MIT
Stars: 369
Forks: 57
Downloads:
Commits (30d): 0
Language: Jupyter Notebook
License: MIT
No Package No Dependents
Stale 6m No Package No Dependents

About PINA

mathLab/PINA

Physics-Informed Neural networks for Advanced modeling

Builds on PyTorch, PyTorch Lightning, and PyTorch Geometric to provide modular Problem, Model, Solver, and Trainer APIs for both supervised learning and physics-informed tasks. Supports Neural Operators and graph-based architectures, with automatic differentiation for constraint enforcement (e.g., differential equations, boundary conditions) and multi-device training via PyTorch Lightning's distributed backend.

About PINN

nanditadoloi/PINN

Simple PyTorch Implementation of Physics Informed Neural Network (PINN)

Scores updated daily from GitHub, PyPI, and npm data. How scores work