Model for mptrj_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 Materials Project trajectory (MPtrj) dataset. Here we use mean absolute error (MAE) vs DFT (PBE). External links: https://figshare.com/articles/dataset/23713842
Reference(s): https://doi.org/10.1038/s42256-023-00716-3, https://github.com/atomgptlab/alignn
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| alignnff_mptrj_2026 | mptrj | 0.051 | AtomGPTLab | 76485 | 07-25-2026 | CSV, JSON, run.sh, Info |
| alignn2_radius | mptrj | 0.051 | AtomGPTLab | 76485 | 07-30-2026 | CSV, JSON, run.sh, Info |