OBJECTIVE Both short and long sleep duration have frequently been found to be associated with an increased risk for diabetes. The aim of the present exploratory analysis was to examine the association between sleep duration and type 2 diabetes after lifestyle intervention in overweight individuals with impaired glucose tolerance in a 7-year prospective follow-up. RESEARCH DESIGN AND METHODS A total of 522 individuals (aged 40–64 years) were randomly allocated either to an intensive diet-exercise counseling group or to a control group. Diabetes incidence during follow-up was calculated according to sleep duration at baseline. Sleep duration was obtained for a 24-h period. Physical activity, dietary intakes, body weight, and immune mediators (C-reactive protein and interleukin-6) were measured. RESULTS Interaction between sleep duration and treatment group was statistically significant (P = 0.003). In the control group, the adjusted hazard ratios (HRs) (95% CI) for diabetes were 2.29 (1.38–3.80) and 2.74 (1.67–4.50) in the sleep duration groups 9–9.5 h and ≥10 h, respectively, compared with for that of the 7–8.5 h group. In contrast, sleep duration did not influence the incidence of diabetes in the intervention group; for sleep duration groups 9–9.5 h and ≥10 h, the adjusted HRs (95% CI) were 1.10 (0.60–2.01) and 0.73 (0.34–1.56), respectively, compared with that in the reference group (7–8.5 h sleep). Lifestyle intervention resulted in similar improvement in body weight, insulin sensitivity, and immune mediator levels regardless of sleep duration. CONCLUSIONS Long sleep duration is associated with increased type 2 diabetes risk. Lifestyle intervention with the aim of weight reduction, healthy diet, and increased physical activity may ameliorate some of this excess risk.