This study evaluates the association between various anthropometric indices and prediabetes prevalence among adults in the United States. We conducted a cross-sectional analysis on participants aged ≥20 years using data from the National Health and Nutrition Examination Survey (NHANES) cycles 2021–2023. Weight, height, waist circumference, and hip circumference were used to calculate indices such as BMI, waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), conicity index, abdominal volume index (AVI), body roundness index (BRI), body adiposity index (BAI), and a body shape index (ABSI). Multivariable logistic regression assessed associations between indices and prediabetes risk. Receiver Operating Characteristic (ROC) curve analysis evaluated the discriminatory power of each index. Among 2515 participants, 49.6 % had prediabetes. Higher quartiles of BMI, Waist circumference, WHR, WHtR, CI, AVI, BRI, BAI, and ABSI were significantly associated with increased prediabetes risk (P < 0.05). WHR demonstrated the highest discriminatory power (AUC: 0.695; 95 % CI; 0.674, 0.716), followed by conicity index (AUC; 0.693; 95 % CI; 0.672, 0.714). Central adiposity measures, particularly WHR and CI, are more effective than BMI in identifying individuals at risk for prediabetes. However, due to the cross-sectional nature of our study, causal relationships between anthropometric indices and prediabetes risk cannot be established.