Stars
Self-describing sparse tensor data format for atomistic machine learning and beyond
Chemical intuition for surface science in a package.
Intelligent Scientific Agent for Materials Design
CBE 504: Chemical Kinetics and Reaction Engineering
A RL framework for Crystal Structure Generation using GRPO
Kosmos: An AI Scientist for Autonomous Discovery - An implementation and adaptation to be driven by Claude Code or API - Based on the Kosmos AI Paper - https://arxiv.org/abs/2511.02824
Space group equivariant crystal diffusion [NeurIPS 2025]
Training Neural Network potentials through customizable routines in JAX.
Public releases of the Clinamen2 optimization framework
Python software that implements the formulation for evaluating the effects of electron-phonon coupling on the optical properties of defects. Interfaces with the mace machine learning interatomic po…
A unified package for post-processing optical properties of point defects from first principles calculation.
mdi-group / mace-field
Forked from ACEsuit/maceMACE-Field - Fast and accurate machine learning interatomic potentials under an external electric field for predicting dielectric properties.
Graph neural network prediction of electronic Hamiltonians in atomic orbital representation with many body messages
Interactive browser visualizations for materials science: periodic tables, 3d crystal structures, MD trajectories, heatmaps, scatter plots, histograms.
Intuitive, easy CLIs based on python type hints.
The architector python package - for 3D metal complex design. C22085
Automated Quantum Mechanical Environments (AQME) offers transparent and reproducible workflows available for Jupyter Notebooks and command lines, including: 1) RDKit- and CREST-based conformer gene…
A machine learning package using the equinox library for learning XC functionals with JAX.
Free energy calculators by Bayesian-inspired nested sampling and other integration techniques
Repository for the Imperial 2023 large language model hackathon.