Diabetes is one of the major global public health problems and is gradually getting worse particularly in developing nations where 95% of patients are suffering from type 2 diabetes (T2D). Animal models in diabetes research are very common where rodents are the best choice of use due to being smaller in size, easy to handle, omnivorous in nature, and non-wild tranquil behavior. Normally rodent models are classified into two major classes namely: (1) genetic or spontaneously induced models and (2) non-genetic or experimentally induced models. Non-genetic models are more popular compared to genetic models due to lower cost, wider availability, easier to induce diabetes, and of course easier to maintain compared to genetic models. A number of non-genetic models have been developed in last three decades for diabetes research including adult alloxan/streptozotocin (STZ) models, partial pancreatectomy model, high-fat (HF) diet-fed models, fructose-fed models, HF diet-fed STZ models, nicotinamide–STZ models, monosodium-glutamate (MSG) induced models, and intrauterine growth retardation (IUGR) models. A T2D model should have the all major pathogenesis of the disease usually found in humans; however, none of the above-mentioned models are without limitations. This chapter comparatively evaluates most of the experimentally induced rodent models of T2D with their limitations, advantages, disadvantages, and criticality of development in order to help diabetes research groups to more appropriately select the animal models to work on their specific research question.