Tea is the most ancient and popular beverage in the world, and its beneficial health effects has attracted tremendous attention worldwide. However, the prospective evidence relating green tea consumption to total and cause-specific mortality is still limited and inconclusive. We recruited 164,681 male participants free of pre-existing disease during 1990–1991, with green tea consumption and other covariates assessed by the standardized questionnaire and mortality follow up continued until 2006 (mean 11 years; total person-years: 1,961,791). Cox regression analyses were used to quantify the associations of green tea consumption with all-cause (n = 32,700), CVD (n = 11,839) and cancer (n = 7002) mortality, adjusting simultaneously for potential confounders. At baseline, 18 % reported regular consumption of green tea. Compared with non-green tea drinkers, regular drinkers had significantly lower all-cause mortality, with adjusted hazard ratios (HRs) being 0.94 (95 % CI 0.89, 0.99) for ≤5 g/day, 0.95 (0.91, 0.99) for 5–10 g/day and 0.89 (0.85, 0.93) for >10 g/day. For CVD mortality, the corresponding HRs were 0.93 (0.85, 1.01) 0.91 (0.85, 0.98) and 0.86 (0.79, 0.93), respectively, while for cancer they were 0.86 (0.78, 0.98), 0.92 (0.83, 1.00) and 0.79 (0.71, 0.88), respectively. The patterns of these associations varied by smoking, alcohol drinking and locality. This large prospective study shows that regular green tea consumption is associated with significantly reduced risk of death from all-cause, CVD and cancer among Chinese adults.