Streptozotocin (STZ)-induced diabetes in ICR mice is often used to model diabetes mellitus and its complications, as well as other pathologies. In studies of diabetes progression and effects of newly developed treatments, experimental results may be difficult to interpret because blood glucose levels (BGLs) of untreated diabetic control animals tend to decline substantially during typical experimental time spans of 8–11 h. To address this problem, the authors examined several experimental conditions that might affect BGL stability, including STZ dose, initial mouse weight, fasting regimen and light:dark cycle. The authors found that diabetes severity was dependent on initial mouse weight and that weight loss after diabetes induction was less severe in heavier mice. Furthermore, a dose of 150 mg STZ per kg body weight was sufficient to induce stabilized acute diabetes without causing many complications. Finally, BGL could be stabilized in diabetic mice that were not treated with insulin by avoiding pre-fasting before an 8-h experiment and by allowing mice limited access to food during the experiment.