The streptozotocin (STZ)-induced diabetic mouse model has been extensively used as a model for diabetes and diabetic nephropathy, but it is still influenced by many off-target toxic effects and large variation in diabetes induction. Therefore, the aim of this study was to compare different STZ dosing regimens to optimise animal welfare and minimise unwanted effects of STZ measured by acute renal toxicity, impairment of stomach emptying and weight loss. Male 129/Sv mice were injected with 1 × 50, 1 × 100, 1 × 125, 1 × 150, 1 × 200, 5 × 50, 2 × 100 and 2 × 125 mg/kg STZ or vehicle and euthanized 24 hours after the last injection. All STZ doses were found to induce significant enlargement of the stomach. All multiple doses of STZ increased the albumin:creatinine ratio significantly, and immunohistochemical staining of KIM-1 and Ki-67 was increased by 5 × 50 and 2 × 100 mg/kg STZ. Renal gene expression of Cdkn1a, KIM-1, NGAL and MCP-1 was increased by most of the STZ doses. No difference was found between the double intermediate dose of 2 × 100 mg/kg and the multiple low dose of 5 × 50 mg/kg regarding either stomach enlargement or kidney injury. However, the reduced fasting periods and injections in the 2 × 100 mg/kg STZ group could have lowered the impact on the general condition measured as change in body weight. This shows that the double intermediate dose is a good alternative to the recommended multiple low dose for diabetes induction in these mice. The STZ-induced mouse model has again proven to be a model with large variations affecting both animal welfare and model robustness.