BackgroundPrevious studies have linked single metal to hemoglobin levels in children and adolescents; however, studies with regards to metal mixtures are still limited. ObjectiveWe aimed to investigate the associations of single metal and metal mixtures with hemoglobin levels in children and adolescents. MethodsWe conducted a cross-sectional study of 2064 children and adolescents aged 6 to 19 years in Liuzhou, China in 2018. The concentrations of 15 metals in urine were determined by inductively coupled plasma mass spectrometry. Generalized linear regression and weighted quantile sum (WQS) regression were used to estimate the associations of single metal and metal mixtures with hemoglobin levels, respectively. ResultsThe multivariable-adjusted β-values for the highest versus the first quartiles of urinary metal concentrations were − 1.57 (95 % confidence interval [CI]: −3.01, −0.13) for chromium, −2.47 (95 % CI: −3.90, −1.05) for nickel and 1.88 (95 % CI: 0.49, 3.28) for copper. In addition, we found a significant negative association between the WQS index and hemoglobin levels (adjusted β = −0.93, 95 % CI: −1.69, −0.19), with nickel contributing the most to the WQS index at 59.0 %. Subgroup analyses showed that exposure to urinary nickel or metal mixtures were associated with decreased hemoglobin levels in adolescents, but not in children (all Pinteration < 0.001). ConclusionAmong children and adolescents, urinary chromium and nickel concentrations were associated with decreased hemoglobin levels, while copper showed a positive relationship. Moreover, a negative association was observed between exposure to metal mixtures and hemoglobin levels. These findings need to be further validated in prospective studies.