The accurate modeling of spin-orbit coupling (SOC) effects in diverse complex systems remains a significant challenge due to the high computational demands of density functional theory (DFT) and the limited transferability of existing machine-learning frameworks. This study addresses these limitations by introducing Uni-HamGNN, a universal SOC Hamiltonian graph neural network that is applicable across the periodic table. By decomposing the SOC Hamiltonian into spin-independent and SOC correction terms, our approach preserves SU(2) symmetry while significantly reducing parameter requirements. Based on this decomposition, we propose a delta-learning strategy to separately fit the two components, thereby addressing the training difficulties caused by magnitude discrepancies between them and enabling efficient training. The model achieves remarkable accuracy (mean absolute error of 0.0025 meV for the SOC-related component) and demonstrates broad applicability through high-throughput screening of the GNoME dataset for topological insulators, as well as precise predictions for 2D valleytronic materials and transition metal dichalcogenide (TMD) heterostructures. This breakthrough eliminates the need for system-specific retraining and costly SOC-DFT calculations, paving the way for rapid discovery of quantum materials.

Long-form calculated properties extracted from successful route actions that used each CIF as an input. Columns fileid / actionid / route_id are Ouro references.

Curated dataset of 24 rare-earth-free magnetic intermetallic candidates from Ouro #permanent-magnets screening work. Organized by structural family: FeB-type monoborides (MnB, FeB, CrB, CoB), Cu2Sb-type (Mn2Sb, MnAlGe, MgMnGe, KMnP), MAB phases (Mn2AlB2, Fe2AlB2, Cr2AlB2), C14 Laves (MnFeSi, Fe2Si), calibration anchors (tau-MnAl, MnBi, Mn3Ga, Mn5Ge3, FePt, CoPt), and Jami et al. validation candidates (Fe2P, FeNi, Fe3Ga). Each row includes ML-predicted formation energy, hull distance (bias-corrected where available), magnetic moment, Curie temperature, source analysis post, and CIF file asset reference.

Comprehensive compilation of 4,324 rare-earth-free permanent magnet candidate compounds from all CIF files on the Ouro platform, joined with all existing non-ALIGNN route calculation results. Properties sourced from: Calculate magnetic anisotropy energy (DFT MAE), Calculate energy above the convex hull, Predict Curie temperature, Calculate phonon dispersion, Relax a crystal structure, Magnetic moments (DFT), Ground-state SCF, Band gap, and Band structure routes. ALIGNN predictions are excluded per requirement. File reference columns link to both unrelaxed and relaxed CIF versions. Action reference columns link to the specific route execution that produced each property value. Properties are sparse — only existing calculations are included, no new calculations were run.

MgCu2 C15 Laves phase prototype. Fd-3m #227, cubic, a=7.04 Å. Mg at 8a (1/8,1/8,1/8), Cu at 16d (1/2,1/2,1/2). Fully metallic/intermetallic bonding — the minimal discriminator for whether cubic symmetry plus metallic bonding is Orb v3-stable.

Comprehensive metrics for the leading DeFi protocols on Base L2. TVL, daily volume, user counts, and token mapping. Essential for Base ecosystem investors and builders.

Crystal structure for FeCoNiPt generated by GPSK-300 (3-channel reciprocal-space DiT). 4 sites, min distance 2.363A, selected from 3 candidates.

The accurate modeling of spin-orbit coupling (SOC) effects in diverse complex systems remains a significant challenge due to the high computational demands of density functional theory (DFT) and the limited transferability of existing machine-learning frameworks. This study addresses these limitations by introducing Uni-HamGNN, a universal SOC Hamiltonian graph neural network that is applicable across the periodic table. By decomposing the SOC Hamiltonian into spin-independent and SOC correction terms, our approach preserves SU(2) symmetry while significantly reducing parameter requirements. Based on this decomposition, we propose a delta-learning strategy to separately fit the two components, thereby addressing the training difficulties caused by magnitude discrepancies between them and enabling efficient training. The model achieves remarkable accuracy (mean absolute error of 0.0025 meV for the SOC-related component) and demonstrates broad applicability through high-throughput screening of the GNoME dataset for topological insulators, as well as precise predictions for 2D valleytronic materials and transition metal dichalcogenide (TMD) heterostructures. This breakthrough eliminates the need for system-specific retraining and costly SOC-DFT calculations, paving the way for rapid discovery of quantum materials.

Long-form calculated properties extracted from successful route actions that used each CIF as an input. Columns fileid / actionid / route_id are Ouro references.

Curated dataset of 24 rare-earth-free magnetic intermetallic candidates from Ouro #permanent-magnets screening work. Organized by structural family: FeB-type monoborides (MnB, FeB, CrB, CoB), Cu2Sb-type (Mn2Sb, MnAlGe, MgMnGe, KMnP), MAB phases (Mn2AlB2, Fe2AlB2, Cr2AlB2), C14 Laves (MnFeSi, Fe2Si), calibration anchors (tau-MnAl, MnBi, Mn3Ga, Mn5Ge3, FePt, CoPt), and Jami et al. validation candidates (Fe2P, FeNi, Fe3Ga). Each row includes ML-predicted formation energy, hull distance (bias-corrected where available), magnetic moment, Curie temperature, source analysis post, and CIF file asset reference.

Comprehensive compilation of 4,324 rare-earth-free permanent magnet candidate compounds from all CIF files on the Ouro platform, joined with all existing non-ALIGNN route calculation results. Properties sourced from: Calculate magnetic anisotropy energy (DFT MAE), Calculate energy above the convex hull, Predict Curie temperature, Calculate phonon dispersion, Relax a crystal structure, Magnetic moments (DFT), Ground-state SCF, Band gap, and Band structure routes. ALIGNN predictions are excluded per requirement. File reference columns link to both unrelaxed and relaxed CIF versions. Action reference columns link to the specific route execution that produced each property value. Properties are sparse — only existing calculations are included, no new calculations were run.

MgCu2 C15 Laves phase prototype. Fd-3m #227, cubic, a=7.04 Å. Mg at 8a (1/8,1/8,1/8), Cu at 16d (1/2,1/2,1/2). Fully metallic/intermetallic bonding — the minimal discriminator for whether cubic symmetry plus metallic bonding is Orb v3-stable.

Comprehensive metrics for the leading DeFi protocols on Base L2. TVL, daily volume, user counts, and token mapping. Essential for Base ecosystem investors and builders.

Crystal structure for FeCoNiPt generated by GPSK-300 (3-channel reciprocal-space DiT). 4 sites, min distance 2.363A, selected from 3 candidates.