Objective To investigate the effect of different postprandial exercise choices on blood glucose of patients with type 2 diabetes mellitus. Methods A total of 120 patients with type 2 diabetes mellitus were enrolled and randomized into 3 groups,who were instructed to go for a brisk walking of 2 km distance at a pace of 4.0-5.0 km/h either 90 min（group A）, 120 min （group B） or 150 min （group C） after breakfast. Glucose levels at fasting, before exercise, immediately after exercise, and at 11 am （before next meal） were measured. Results There were no significant differences in the levels of fasting and pre-exercise blood glucose among the three groups （P＞0.05 for both）.The blood glucose levels immediately after exercise, and at 11 am had significant differences among the three groups （P＜0.05,P＜0.01）, with the comparisons of group B versus group A, and group C versus group A, presenting significant differences （P＜0.05,P＜0.01）. At 11 am, there were 5 cases developing hypoglycemia in group C. Conclusion Doing exercises 120 min and 150 min postprandially are superior to exercises 90 min postprandially regarding glucose reduction in patients with type 2 diabetes mellitus. However, given that doing exercises 150 min postprandially could yield higher risk of hypoglycemia, a postprandial 120 min scheme is more favorable, and still needs further testing.