Abstract
Conjugated organic polymers are the cornerstone of modern organic electronics, yet accurate prediction of their properties remains a challenging task due to their synthet-ic complexity and high computational cost of quantum-chemical methods. Here, we develop a graph neural network based on the DimeNet++ direct message passing ar-chitecture to predict HOMO, LUMO and band gap energies of conjugated polymers directly from their 3D monomer structure. The model was pre-trained on TD-DFT-extrapolated data and trained on a limited dataset of experimentally meas-ured properties. As a result, pre-training had significantly improved model’s accuracy compared to direct training (MAEs ~0.3 eV vs 0.074 eV, 0.141 and 0.172 for HO-MO/LUMO and band gap, respectively). Pre-training on monomer DFT data did not provide comparable gains. The results demonstrate that polymer-relevant pretraining is critical for capturing structure–property relationships and enables accurate predic-tions without delta-learning or prior quantum-chemical calculations, facilitating effi-cient screening and rational design of conjugated polymers for organic optoelectronics.