Table 2.

Datasets used in the combined dataset

Dataset	Number of SMILES
ZINC 15 [31]	5M
QM9 [32, 33]	134k
ZINC 250k [34]	250k
RedDB [35]	31k
OPV [36]	91k
PubchemQC 2017/2020 [37, 38]	5.3M
CEP [39] subset [40]	20k
ChEMBL [41–44]	2.3M
Combined (OrganiX13)	13.1M (Total) / 12.5M (After removing duplicates)