Several different rodent models are available for metabolic studies on the development of diabetes. Although the abnormalities associated with each diabetes type have many features in common, the documentation of several different genes being involved makes it unlikely that the various syndromes will be reduced to a single disturbance in one metabolic pathway. The severity of the diabetes produced depends on the interaction of the individual mutation with genetic factors in the inbred background of the host. Establishing the nature of these gene-host interactions in rodents should aid us in understanding similar interactions that occur in human diabetes. The development of the syndrome in most models is similar and includes hyperinsulinemia, hyperphagia, and'attempts at increasing insulin supply by β-cell hyperpiasia and hypertrophy in the early stages. Hyperglycemia, obesity, and severe diabetes are secondary features that result from a combination of insulin resistance and a failure to sustain the secretion of the large amounts of insulin. Most models utilize ingested food and stored food reserves more efficiently. This increased metabolic efficiency extends to heterozygotes that are normal in all respects having only one dose of the deleterious gene. Establishing this increased metabolic efficiency in heterozygotes lends credence to the thrifty gene hypothesis of diabetes and suggests a mechanism whereby some deleterious diabetes genes may be favored in the human population. The best studied mouse models, and those for which the most complete information is available, are those caused by single genes, e.g., yellow, obese, diabetes, tubby, and fat. In the other models, the mode of inheritance is either polygenic or otherwise unclear, features which interfere with the interpretation of the data. This report briefly summarizes the developing syndrome in each model, points out any differences, and suggests the most appropriate areas where future research should be most productive in the light of contemporary studies.