The aim of the present study was to develop an alternative non-obese non-genetic rat model of type 2 diabetes (T2D). Six-week-old male SD rats were randomly divided into six groups, namely: Normal Control (NC), Diabetic Control (DBC), Caffeine 5mg/kg BW+STZ (CAF5), Caffeine 10mg/kg BW+STZ (CAF10), Caffeine 20mg/kg BW+STZ (CAF20) and Caffeine 40mg/kg BW+STZ (CAF40) and were fed a normal rat pellet diet and drinking water ad libitum throughout the experimental period. After a one week acclimatization period, diabetes was induced in the animals in DBC and all CAF groups with an injection (i.p.) of the respective dosages of caffeine (mg/kg BW) 15min before the injection of STZ (65mg/kg BW) when normal saline was injected to the DBC group instead of caffeine. The NC group received normal saline and buffer instead of caffeine and STZ, respectively. One week after the STZ injection, animals with non-fasting blood glucose>300mg/dl were considered as diabetic. Three weeks after the STZ injection, the animals in the CAF5 and CAF10 groups were eliminated from the study due to the severity of diabetes and the experiment was continued with the remainder groups for a 13 weeks period. The data of food and fluid intake, body weight, blood glucose, glucose tolerance test, HOMA-IR, HOMA-beta, serum insulin, fructosamine, lipid profile and organ specific enzymes, anti-diabetic drug response tests, and pancreatic histopathology suggest that CAF20 group can be a better alternative non-genetic model of non-obese T2D.