This study aims to provide the initial validity of the Autism Diagnostic Observation Schedule-2 (ADOS-2) Toddler Module and Module 1-2 for South Korean toddlers and preschoolers. Based on 143 children, the ASD group (n = 68) showed significantly higher ADOS-2 item and algorithm total scores as well as social affect and repetitive and restricted behaviors domain scores compared with children with nonspectrum (NS; n = 42) disorders and typically developing (TD; n = 33) children. Using lower algorithm cutoffs, sensitivities were excellent for the ASD versus NS/TD comparisons, ranging from 94% to 100% across different Modules. Specificities varied more, ranging from 82% to 100%. Internal consistency was strong with high item-total correlations (r of 0.6-0.9) and Cronbach's Alphas (all above 0.7). Results demonstrated promising, initial evidence for the validity of the ADOS-2 for South Korean toddlers and preschoolers from 1 to 4 years of age. The ADOS-2 could be implemented, with minimal adaptations, in research and clinical settings in South Korea. This study is one of the first steps toward validating the ADOS-2 in other Eastern countries that are in great need for a valid instrument for the detection of ASD. Autism Res 2019, 12: 1356-1366. © 2019 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: Results of this study demonstrated promising, initial evidence for the validity of a gold standard measure for the diagnosis of autism, the Autism Diagnostic Observation Schedule-2 (ADOS-2), for South Korean toddlers and preschoolers. The ADOS-2 could be implemented, with minimal adaptations, in research and clinical settings in South Korea. This study is one of the first steps toward validating the ADOS-2 in other Eastern countries that are in great need of a valid instrument for the detection of ASD.