Building Ouro, using AI to search for room-temp superconductors and rare-earth free permanent magnets.
Predicts adsorption energy from the Open Catalyst Project 2020 (100k subset).
Predicts the electronic (high-frequency) contribution to the dielectric function ε∞x.
Predicts the maximum component of the dielectric tensor from DFPT calculations.
Predicts the maximum piezoelectric strain coefficient dij from DFPT calculations.
Predicts the Voigt-averaged bulk modulus Kv.
Predicts the Voigt-averaged shear modulus Gv.
Predicts the exfoliation energy, useful for identifying cleavable 2D materials.
Predicts the n-type Seebeck coefficient at 600 K from the JARVIS-DFT dataset.
Predicts the p-type Seebeck coefficient at 600 K from the JARVIS-DFT dataset.
Predicts the n-type thermoelectric power factor.
Predicts nitrogen adsorption energy on metal surfaces using the TinNet dataset.
Predicts the HOMO-LUMO gap for molecular structures (QM9 dataset).
Predicts the superconducting critical temperature Tc.
Predicts the electronic density of states at the Fermi level for superconductor analysis.
Predicts the Debye temperature for superconductor analysis.
Predicts the Eliashberg spectral function α²F(ω) sampled at 100 frequency points.
Predicts the phonon density of states sampled at 66 frequency points.
Predicts the optimal k-point length unit for DFT convergence studies.
Predicts oxygen adsorption energy on metal surfaces using the TinNet dataset.
Predicts free energy at 298.15 K for molecular structures (QM9 dataset).