BACKGROUND: The worldwide incidence of non-alcoholic fatty liver disease (NAFLD) is a pressing health issue, with chronic inflammation critically influencing its development. The Dietary Inflammatory Index (DII) measures the inflammatory potential of one's diet. Our study aimed to investigate the association between DII and fatty liver disease in obese and non-obese individuals. METHODS: Data from the 1999--2001 National Health and Nutrition Examination Survey (NHANES), encompassing 3456 adults aged 20 years and above, were utilized for the analysis. The multivariable analysis assessed the correlation between DII, NAFLD, and obesity while adjusting for demographic variables. The restricted cubic spline (RCS) method was also applied with restricted degrees of freedom to establish a non-linear relationship model between DII and NAFLD. RESULTS: Our study demonstrated significant differences across DII tertiles in various demographic and clinical characteristics. An increase in t DII score by one standard deviation resulted in a 21% increase in the fatty liver risk. This association persisted after adjusting for age, sex, and other covariates. The risk of fatty liver disease increased significantly by 39% in the highest DII tertile. Regression analysis revealed a non-linear correlation between DII and the risk of fatty liver in obese adults. The subgroup analysis showed consistent findings across all subgroups, with particularly strong associations in specific demographic categories. CONCLUSIONS: Our findings suggest a significant association between the DII score and risk of fatty liver disease, particularly in obese individuals. These results underscore the potential role of diet-induced inflammation in pathogenesis of NAFLD.