OBJECTIVE To examine associations between specific types of physical activity and all-cause and cardiovascular disease (CVD) mortality in a large nationally representative sample of adults with diabetes from Great Britain. RESEARCH DESIGN AND METHODS There were a total of 3,038 participants (675 deaths) with diabetes in the Health Survey for England and the Scottish Health Surveys conducted between 1997 and 2008. Participants aged ≥50 years at baseline were followed up for an average of 75.2 months for all-cause and CVD mortality. Data were collected on self-reported frequency, duration, and intensity of participation in sports and exercise, walking, and domestic physical activity, from which the number of MET-hours/week were derived. Sex-specific medians of time spent in each type of physical activity (for those physically active) were calculated, and Cox proportional hazards regression conducted to examine type-specific associations between the level of physical activity and all-cause and CVD mortality risk. RESULTS Inverse associations with all-cause and CVD mortality were observed for overall physical activity in a dose-response manner after adjusting for covariates. Compared with those who individuals were inactive, participants who reported some activity, but below the recommended amount, or who met the physical activity recommendations had a 26% (95% CI 39–11) and 35% (95% CI 47–21) lower all-cause mortality, respectively. Similar results were found for below/above median physical activity levels. Sports and exercise participation was inversely associated with all-cause (but not CVD) mortality, as were above average levels of walking. Domestic physical activity was not associated with mortality. CONCLUSIONS Moderate physical activity levels were associated with better prognosis in diabetic adults.