Model for matpes_energy¶
- Description: This is an AI benchmark to evaluate how accurately a machine learning force-field (MLFF) can predict the total energy (per atom) of the MATPES dataset (single-point PBE energies, forces and stresses sampled from molecular-dynamics trajectories across the periodic table). Here we use mean absolute error (MAE) to compare MLFFs with respect to DFT (PBE) accuracy. External links: https://matpes.ai
Reference(s): https://github.com/atomgptlab/alignn, https://doi.org/10.48550/arXiv.2503.04070
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| alignn2_radius | matpes | 0.0518 | AtomGPTLab | 434712 | 07-30-2026 | CSV, JSON, run.sh, Info |
| alignnff_matpes | matpes | 0.047 | AtomGPTLab | 434712 | 07-24-2026 | CSV, JSON, run.sh, Info |