Model for oqmd_ltc¶
- Description: This is a benchmark to evaluate how accurately an AI model can predict the log10 lattice thermal conductivity (LTC, W/mK) of inorganic crystals. Structures are from OQMD; DFT LTC values from Ojih et al. high-throughput BTE calculations. Target is log10(LTC) spanning ~5.5 orders of magnitude. MAE (log10 units) vs DFT. External links: https://doi.org/10.1039/D3TA04874H
Reference(s): https://doi.org/10.1038/npjcompumats.2015.10, https://github.com/atomgptlab/alignn
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| alignn_2026_ltc_knn | oqmd | 0.3753 | AtomGPTLab | 4034 | 07-26-2026 | CSV, JSON, run.sh, Info |
| alignn_2026_ltc_radius | oqmd | 0.3861 | AtomGPTLab | 4034 | 07-26-2026 | CSV, JSON, run.sh, Info |
| alignn_2026_ltc_dgl | oqmd | 0.3705 | AtomGPTLab | 4034 | 07-26-2026 | CSV, JSON, run.sh, Info |