Abstract Aims/hypothesis The association between pubertal timing and type 2 diabetes, independent of prepubertal BMI, is not fully understood. The aim of the present study was to evaluate the association between pubertal timing and risk of adult type 2 diabetes, independent of prepubertal BMI, in Swedish men. Methods We included 30,697 men who had data for BMI at age 8 and 20 years and age at Peak Height Velocity (PHV), an objective assessment of pubertal timing, available from the BMI Epidemiology Study Gothenburg (BEST Gothenburg), Sweden. Information on type 2 diabetes (n = 1851) was retrieved from the Swedish National Patient Register. HRs and 95% CIs were estimated by Cox regression analysis. We observed violations of the assumption of proportional hazards for the association between age at PHV and the risk of type 2 diabetes and therefore split the follow-up period at the median age of type 2 diabetes diagnosis (57.2 years of age) to define early (≤57.2 years) and late (>57.2 years) type 2 diabetes diagnosis. Results Age at PHV was inversely associated with both early (HR 1.28 per year decrease in age at PHV, 95% CI 1.21, 1.36) and late (HR 1.13, 95% CI 1.06, 1.19) type 2 diabetes. After adjustment for childhood BMI, the associations between age at PHV and both early (HR 1.24, 95% CI 1.17, 1.31) and late (HR 1.11, 95% CI 1.05, 1.17) type 2 diabetes were similar. Moreover, early age at PHV predicted insulin treatment of type 2 diabetes (OR 1.25 per year decrease in age at PHV, 95% CI 1.17, 1.33). Assuming a higher risk among those with an age at PHV below the median, the population attributable factor indicates that 15% fewer of the diagnosed individuals would have developed type 2 diabetes had they not reached puberty early. Conclusions/interpretation These findings indicate that early puberty may be a novel independent risk factor for type 2 diabetes.