Predict material properties
Estimate physical properties of an atomic structure from its atoms and geometry.
Desired outcome
Numerical property estimates for a supplied atomic structure.
Explore from the outcome
Where this function can lead
These documented connections are starting points for selection, not a ranked list.
Model releases
Identifiable models documented for this function.
- IBM pos-egnn v1-6M
Predicts energy, per-atom forces and stress for atomic structures using a position-based graph neural network.
Link evidence: IBM Research materials.pos-egnn model card, IBM pos-egnn example notebook
Typical inputs
- Atom types, positions, and structural context
Typical outputs
- Predicted energy, forces, or stress values
Sources
Last reviewed 2026-09-30