Purposes Previous studies show inconsistent associations between niacin supplementation and diabetes mellitus (DM) in high-risk population, but little is known about the relationship between dietary intake of niacin and DM in the generation population. Our study aimed to explore the associations of dietary niacin intake with the risk of DM in the United States (US) adult population. Methods These data were derived from the National Health and Nutrition Examination Survey (NHANES) 2003-2018 of 35,606 individuals aged 20 years or older. Niacin intake and food sources were measured by two 24-h dietary recall interviews. The diagnosis of DM was established according to the American Diabetes Association (ADA) criteria. Binary logistic regression and restricted cubic spline models were applied to evaluate the association of dietary niacin intake and DM. Results Among the 35,606 individuals, the prevalence of DM was 11.47%. The full-adjusted odds ratio(aOR) of DM was 1.27(95%CI 1.06-1.52) for quartile (Q) 4 v. Q1 of dietary niacin intake. In the dose-response analysis, the shape of the association of niacin intake with the risk of DM was approximately J-shaped (non-linear, p < 0.05). Energy-adjusted niacin of 26.08 mg/day was the optimal cut-off value for predicting DM. Conclusions High dietary niacin intake was positively associated with DM among US adults.