We are potentially exposed to more than a million hazardous chemicals. Determining which chemicals pose health risk, and at what concentration, requires robust epidemiological evidence and reliable methods for assessing exposure. Biomonitoring studies like NHANES (National Health and Nutrition Examination Survey) demonstrate substantial variabilities in biomarker concentrations, but do not measure external exposures, which could identify sources and inform intervention strategies. This study aimed to assess whether we can use high throughput mechanistic models and omics technologies to identify exposure sources and disparity associated with chemical exposures measured in NHANES. We used continuous data for 541 chemical biomarkers and demographic information collected 1999-2014 for 82,091 participants in NHANES cross-sectional study of the non-institutionalized, civilian US population, as model input parameters. We used PROTEX mechanistic model to estimate external exposure levels that could give rise to the internal levels of measured biomarkers. The PROTEX model was parameterized for estimating exposures for adults and living in ""average"" US environment, with variations in anthropometric parameters such as BMI, fat percent, age, ethnicity, time spent indoors taken from the NHANES datasets. Preliminary results suggest that 95% of brominated and chlorinated biomarkers came from lipid-rich diets, even though they may be intensively used in the indoor environment. Concentrations of these biomarkers were significantly higher in non-Hispanic Blacks and Mexicans compared to non-Hispanic Whites. These results show promise in the use of high throughput mechanistic models as tools to comprehensively assess external chemical exposure, and internal exposures, leveraging existing biomonitoring and survey databases.