Replaces custom-trained interatomic potentials with pre-trained equivariant neural network models. Enables high-accuracy force field predictions across diverse chemical spaces without retraining.
Tech Stack
ShellChemistry & Materials
Why mace-foundations?
• State-of-the-art accuracy
• GPU accelerated
• Pre-trained on MP/OMAT
Limitations
• High VRAM requirements
• Complex dependency chain
• Steep learning curve
3/5/2026
Last Update
21
Forks
3
Issues
MIT
License
Financial Leak Detected
Stop the "SaaS Tax"
Your team could be burning cash. Switching to mace-foundations instantly boosts your runway.
