nequix 0.3.0

Nequix source code

Nequix

See more information in our preprint.

Usage

Installation

pip install nequix

or for torch

pip install nequix[torch]

ASE calculator

Using nequix.calculator.NequixCalculator, you can perform calculations in ASE with a pre-trained Nequix model.

from nequix.calculator import NequixCalculator

atoms = ...
atoms.calc = NequixCalculator("nequix-mp-1", backend="jax")

or if you want to use the faster PyTorch + kernels backend

...
atoms.calc = NequixCalculator("nequix-mp-1", backend="torch")
...

Training

Models are trained with the nequix_train command using a single .yml configuration file:

nequix_train <config>.yml

or for Torch

# Single GPU
uv run nequix/torch/train.py <config>.yml
# Multi-GPU
uv run torchrun --nproc_per_node=<gpus> nequix/torch/train.py <config>.yml

To reproduce the training of Nequix-MP-1, first clone the repo and sync the environment:

git clone https://github.com/atomicarchitects/nequix.git
cd nequix
uv sync

Then download the MPtrj data from https://figshare.com/files/43302033 into data/ then run the following to extract the data:

bash data/download_mptrj.sh

Then start the training run. The first time this is run it will preprocess the data into HDF5 files:

nequix_train configs/nequix-mp-1.yml

This will take less than 125 hours on a single 4 x A100 node. The batch_size in the config is per-device, so you should be able to run this on any number of GPUs (although hyperparameters like learning rate are often sensitive to global batch size, so keep in mind).

Citation

@article{koker2025training,
  title={Training a foundation model for materials on a budget},
  author={Koker, Teddy and Smidt, Tess},
  journal={arXiv preprint arXiv:2508.16067},
  year={2025}
}