Training neural network potentials

GRACE models and gracemaker (as implemented in TensorPotential package)

MatID is a Python package for identifying and analyzing atomistic systems based on their structure.

Structural Activation via Metal Oxidation State Analysis. A solvent removal protocol generating activated crystal structures from experimental crystallographic information.

Site-Occupation Disorder

A foundational potential energy dataset for materials

[NeurIPS 2024] source code for "A Recipe for Charge Density Prediction"

Compilation of code employed in the construction and analysis of the MOSAEC database.

Licensed under Academic Non-Commercial Share-Alike License

MGnify API (v2) and production automation system

A workflow for reproducible and open scientific articles

convolutional MBDF via fast fourier transforms

Official implementation of DeepDFT model

A molecular simulation package integrating MLFFs in MOFs for DAC

A module to obtain and visualize k-vector coefficients and obtain band paths in the Brillouin zone of crystal structures

Official implementation of MatterGen -- a generative model for inorganic materials design across the periodic table that can be fine-tuned to steer the generation towards a wide range of property c…

Deep learning quantum Monte Carlo for electrons in real space

Efficient And Fully Differentiable Extended Tight-Binding

train and use graph-based ML models of potential energy surfaces

MatterSim: A deep learning atomistic model across elements, temperatures and pressures.

Algorithms to analyze and predict molecular structures

A unified platform for fine-tuning atomistic foundation models in chemistry and materials science

Machine-Learned Interatomic Potential eXploration (mlipx) is designed at BASF for evaluating machine-learned interatomic potentials (MLIPs). It offers a growing set of evaluation methods alongside …

Molecular dynamics and Monte Carlo soft matter simulation on GPUs.