Background: Diabetes increases the risk of stroke; however, the additional risk conferred by diabetes may vary. This study investigated the association of diabetes with incident stroke by age, race, and sex subgroups among participants in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study. Methods: This study included 23,002 black and white adults without prevalent stroke at baseline (2003-2007). Diabetes was defined as fasting glucose ≥ 126mg/dL, random glucose ≥ 200mg/dL or use of glucose-lowering medication. Incident stroke events were adjudicated and available through 9/30/2016. Cox proportional hazards models were used to assess the association of diabetes with stroke by age, race, and sex subgroups. Results: The prevalence of diabetes was 18.4% (27.5% among blacks; 13.7% among whites). During follow-up, 1018 stroke events occurred. Among those aged <65 years, diabetes conferred a higher risk of stroke for white women (HR=3.11; 95% CI= 1.74, 5.57), black women (HR=1.67; 95% CI=1.07, 2.61) and white men (HR=1.78; 95% CI=1.10, 2.89) but not black men (HR = 1.18; 95% CI= 0.71, 1.96). Among those aged ≥65 years, diabetes conferred a higher risk of stroke for white women (HR=1.88; 95% CI=1.30, 2.70) and black men (HR = 1.72; 95% CI=1.15, 2.57), but not black women (HR=1.06, 95% CI=0.75,1.50) or white men (HR=0.91, 95% CI=0.65, 1.27). Conclusion: In the US, the risk of stroke associated with diabetes varied by age, race, and sex. This was particularly evident among those aged <65 years, highlighting the need for targeted prevention strategies in this high-risk population.