Although significant advances are occurring in epilepsy research, about 30% of epileptic patients are still inadequately controlled by standard drug therapy. For this reason, it continues to be important to develop new chemical entities through which epilepsy could be effectively controlled. In this study, the anticonvulsant activity of forty-two dihydropyrimidin(thi)ones was explored and their efficacy was evaluated in rodents against the maximal electroshock seizure (MES) and subcutaneous pentylenetetrazole tests. The results of preliminary pharmacological screening after intraperitoneal injection in mice revealed that twenty-four compounds showed protection in half or more of the animals tested in the MES seizure model, being nine of them active at the lowest dose tested (30 mg/kg). Structurally, the most promising compounds (both urea and thiourea derivatives) presented smaller lateral chains and unsubstituted or para-substituted phenyl ring with a methyl group. Compounds 4, 5 and 11 also protected against MES-induced seizures in 50–75% of rats after oral administration at 30 mg/kg. Moreover, the minimal motor and/or neurological impairment evaluated through the rotarod assay showed that around 52% of the compounds presented lower toxicity than the antiepileptic drugs lamotrigine, carbamazepine and phenytoin. In addition, the most active compounds did not show notable cytotoxicity in in vitro experiments conducted in several cell lines (relative cell proliferation higher than 50% at 30 μM), which can be relevant due to the fact that the toxicity is a common problem of the available antiepileptic drugs. Furthermore, additional computational studies indicated that all compounds respected the Lipinski's rule of five, which, together with the data of efficacy and toxicity, make them attractive compounds to be developed in the future as potential anticonvulsant agents.