According to International Conference on Harmonization guidelines, each drug in development for administration to women of child-bearing potential must be tested for possible developmental toxicities using at least two species (a rodent and non-rodent). With the high cost and slow pace of embryonic-fetal toxicity testing in mammals, both the zebrafish embryonic toxicity test (ZET) and mouse embryonic stem cell test (mEST) have been shown to be useful to assess developmental toxicity of various chemical compounds, including human pharmaceutical drugs, in a high-throughput manner. However, comparative study of the sensitivity and specificity of these methods using the same set of human pharmaceutical drugs is scarce. In this study, we assessed developmental toxicity tests of 39 chemical compounds, including human pharmaceutical drugs, in both the ZET and mEST. The accuracy, sensitivity, and specificity of the ZET were 69%, 59%, and 82%, respectively. The accuracy, sensitivity, and specificity of the mEST were 64%, 50%, and 82%, respectively. As a result, both the ZET and mEST showed acceptable accuracies compared with rat embryo-fetal toxicity study and Food and Drug Administration pregnancy categories. By comparing the results between the ZET and mEST, we identified different types of true positives and true negatives. Thus, complementary tests using both the ZET and mEST may better predict the developmental toxicity of human pharmaceuticals.