| --- |
| library_name: matgl |
| tags: |
| - matgl |
| - materials-science |
| - graph-neural-network |
| - machine-learning-interatomic-potential |
| - foundation-potential |
| - mlip |
| --- |
| |
| ## Introduction |
|
|
| Pre-trained TensorNet foundation potential, i.e., universal machine learning interatomic potential trained on the MatPES r2SCAN 2025.2 dataset. |
|
|
| ## Potential |
|
|
| [matgl](https://github.com/materialsvirtuallab/matgl) `Potential` model (version 3). |
|
|
| ## Usage |
|
|
| ```python |
| import matgl |
| |
| model = matgl.load_model("materialyze/TensorNet-PES-MatPES-r2SCAN-2025.2") |
| ``` |
|
|
| ## Stats |
| - Layers: 2 |
| - Units: 128 |
| - Test_MAE_energies: 32 meV/atom |
| - Test_MAE_forces: 142 meV/Å |
| - Test_MAE_stresses: 0.705 GPa |
|
|
| ## Metadata |
|
|
| ```json |
| { |
| "dataset": "MatPES-r2SCAN-2025.2" |
| } |
| ``` |
|
|
