The analogue insulin glulisine (Glu) shows both more rapid onset and shorter duration of action compared with the other rapid-acting insulin analogues. The current study investigates these properties in regard to the occurrence of hypoglycemia related to exercise. A randomized, single-center, open-label, crossover study was conducted in 12 hospitalized type 2 diabetes patients (all male, mean ± SD age of 51.9 ± 11.3 years; BMI: 25.5 ± 3.9 kg/m2; HbA1c: 11.2 ± 2.4 %). Glu or insulin aspart (Asp) was subcutaneously administered just before breakfast. Insulin dosage was determined as the usual dose of pre-prandial rapid-acting insulin for patients treated with insulin therapy or as 0.1 unit/kg for patients treated with oral anti-hyperglycemic agents. Sixty min after the start of eating, the patients began aerobic exercise on a bicycle ergometer for 30 min at 50% of maximum heart rate. Hypoglycemic episodes (plasma glucose level < 70 mg/dL with or without symptoms) were observed more frequently in Asp group (p < 0.05). Post-exercise plasma glucose levels at 90, 120, and 150 min were significantly lower in Asp group (p < 0.05). In patients with BMI < 25 kg/m2 (n = 6), post-exercise blood glucose levels were significantly lower in Asp group (p < 0.05), while in patients with BMI ≥ 25 kg/m2 (n = 6) the difference was not significant. Glu may therefore be a suitable choice of rapid-acting insulin for patients with type 2 diabetes who are at high risk of post-exercise hypoglycemia.