Diabetes mellitus (DM) is associated with many health complications and is potentially a morbid condition. As prevalence increases at an alarming rate around the world, research into new antidiabetic compounds with different mechanisms is the top priority. Therefore, the preclinical experimental induction of DM is imperative for advancing knowledge, understanding pathogenesis, and developing new drugs. Efforts have been made to examine recent literature on the various induction methods of Type I and Type II DM. The review summarizes the different in vivo models of DM induced by chemical, surgical, and genetic (immunological) manipulations and the use of pathogens such as viruses. For good preclinical assessment, the animal model must exhibit face, predictive, and construct validity. Among all reported models, chemically induced DM with streptozotocin was found to be the most preferred model. However, the purpose of the research and the outcomes to be achieved should be taken into account. This review was aimed at bringing together models, benefits, limitations, species, and strains. It will help the researcher to understand the pathophysiology of DM and to choose appropriate animal models.