NSF NCAR Community Research Earth Digital Intelligence Twin (CREDIT)

FuXi 1.0 degree pressure level model with global mass and energy conservations

This repository contains the PyTorch checkpoint weights, configuration files, and mean/std inputs for the CREDIT FuXi-base and FuXi-physics. More information is available at:

  • Sha, Y., J. Schreck, W. Chapman, D. J. Gagne II, 2025: Improving AI weather prediction models using global mass and energy conservation schemes. arXiv:2501.05648. Pre-print: https://arxiv.org/abs/2501.05648

The GitHub repository of this work is at: https://github.com/NCAR/CREDIT-physics-run

The FuXi models of this repository are modified from the original FuXi and developed at the NSF-National Center for Atmospheric Research (NCAR) for research purposes.

Notes

To run the model, install the CREDIT package and modify the paths in the YAML configuration files.

The datasets of the model will be prepared soon.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.