THEMol: Torsion, Hessian, and Energy Data for Organic Molecules

THEMol is an open-source collection of quantum mechanical properties for organic molecules. It provides large-scale density functional theory (DFT) data, including optimized geometries, structural relaxation trajectories, torsion scans, and Hessian matrices. The dataset was created by ByteDance-Seed and was last updated on 2026-05-15.

Use Cases

• Training machine learning force fields based on Hessian matrices and energy data.
• Analyzing intramolecular potential energy surfaces using torsion scan trajectories.
• Developing models for molecular geometry optimization based on structural relaxation data.
• Predicting atomic properties using MBIS-derived features.

Strengths

• Dataset is described as a large-scale collection of DFT data.
• Includes multiple quantum mechanical property types such as Hessian matrices and torsion scans.
• Open-source and tailored for organic molecules.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

ByteDance-Seed on Hugging Face.

Collection Method

Data generated via density functional theory (DFT) calculations.

Freshness

Last updated 2026-05-15 02:45:44.