Model for halide perovskite PBE bandgap¶
- Description: This is a benchmark to evaluate how accurately an AI model can predict the decomposition energy at the accuracy level of PBE (hybrid DFT functional) using the Halide perovskite dataset of Prof. Arun Mannodi at Purdue. Here we use mean absolute error (MAE) to compare models with respect to DFT (PBE) accuracy. External links: https://github.com/mannodiarun
Reference(s): https://github.com/atomgptlab/jarvis_leaderboard, https://www.nature.com/articles/s41524-021-00650-1, https://github.com/atomgptlab/alignn, https://doi.org/10.1039/D1EE02971A
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| alignn2_radius | halide_peroskites | 0.101 | AtomGPTLab | 229 | 07-30-2026 | CSV, JSON, run.sh, Info |
| alignn_model | halide_peroskites | 0.1223 | ALIGNN | 229 | 01-14-2023 | CSV, JSON, run.sh, Info |
| baseline_mean | halide_peroskites | 0.9895 | AtomGPTLab | 229 | 07-31-2026 | CSV, JSON, run.sh, Info |
| alignn2_knn | halide_peroskites | 0.0867 | AtomGPTLab | 229 | 07-30-2026 | CSV, JSON, run.sh, Info |