#124124th of 126 in ML frameworks and librariesFor science, health and research, science
pinns
Reference code for physics-informed neural networks solving and discovering PDEs
- Stars
- 6.2k
- License
- MIT
- Last commit
- Feb 2026
- Language
- Python
5,693 on 4 Apr 20266,234 on 31 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Original research code from the Raissi, Perdikaris and Karniadakis papers on physics-informed neural networks. It trains networks that respect nonlinear partial differential equations, covering data-driven solution and data-driven discovery of PDEs, in continuous-time and discrete-time forms. The repository is no longer maintained, and the README points to PyTorch, JAX and TensorFlow v2 ports.
Who it is for: Researchers studying PINNs and PDE solving with neural networks
Strengths
- Reference implementation of the 2019 Journal of Computational Physics PINN paper
- Covers both forward solution and inverse discovery of nonlinear PDEs
- Continuous-time and discrete-time model variants are both described
- MIT license with an accompanying project website
Weaknesses
- README states the repository is no longer under active maintenance
- No install instructions, requirements or usage steps in the README
- Research scripts rather than a packaged library; no tagged releases
- Maintained ports exist in PyTorch, JAX and TensorFlow v2
What it needs
- Models: physics-informed neural networks
Quick answers
Is pinns open source?
Yes. Its code is on GitHub under the MIT license.
Also in ML frameworks and libraries
See all 126| Rank | Project | Score |
|---|---|---|
| 1 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 95 out of 100 |
| 2 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |