NVIDIA/physicsnemo

Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods

74
/ 100
Verified

Provides modular PyTorch-compatible building blocks including neural operators (FNOs, DeepONet), graph neural networks, and physics-informed layers, with built-in datapipes for scientific data structures like point clouds and meshes. Features GPU-optimized distributed training utilities that scale from single to multi-node clusters, plus domain-specific sub-modules for CFD, data curation, and weather/climate applications.

2,530 stars. Actively maintained with 49 commits in the last 30 days.

No Package No Dependents
Maintenance 23 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

2,530

Forks

605

Language

Python

License

Apache-2.0

Last pushed

Mar 12, 2026

Commits (30d)

49

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NVIDIA/physicsnemo"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.