Determining the ideal ratio of macronutrients for increasing life expectancy remains a high priority in nutrition research. We aim to investigate the association between carbohydrate, fat, and protein intake and all-cause mortality in Koreans. This cohort study investigated 42,192 participants from the Korea National Health and Nutrition Examination Survey (KNHANES) linked with causes of death data (2007–2015). Hazard ratios (HRs) were calculated using the multivariable Cox proportional regression model after adjusting for confounders. We documented 2110 deaths during the follow-up period. Time to exceed 1% of the all-cause mortality rate was longest in participants with 50–60% carbohydrate, 30–40% fat, and 20–30% protein intake. Adjusted hazard ratio (HR) with 95% confidence intervals (CIs) was 1.313 (1.031–1.672, p = 0.0272) for <50% carbohydrate intake, 1.322 (1.116–1.567, p = 0.0013) for ≥60% carbohydrate intake, 1.439 (1.018–2.035, p = 0.0394) for <30% fat intake, and 3.255 (1.767–5.997, p = 0.0002) for ≥40% fat intake. There was no significant association between protein intake proportion and all-cause mortality. We found a U-shaped association between all-cause mortality and carbohydrate intake as well as fat intake, with minimal risk observed at 50–60% carbohydrate and 30–40% fat intake. Our findings suggest current Korean dietary guidelines should be revised to prolong life expectancy.