Replaces non-equivariant ML models for atomic-scale simulations. Implements Euclidean symmetry-preserving architectures for high-accuracy force and energy prediction.
Tech Stack
PythonChemistry & Materials
Why nequip?
• Extreme data efficiency
• Preserves physical symmetries
• Integrates with ASE and LAMMPS
Limitations
• High computational cost per step
• Complex tensor field theory
• Limited to small-to-medium systems
3/5/2026
Last Update
201
Forks
3
Issues
MIT
License
Financial Leak Detected
Stop the "SaaS Tax"
Your team could be burning cash. Switching to nequip instantly boosts your runway.
