Prenatal vanadium exposure is reported to be associated with restricted fetal growth and adverse birth outcomes. However, trimester-specific vanadium exposure in relation to early-childhood growth still remains unclear. A total of 1873 Chinese mother-infant pairs from whom a complete series of maternal urinary samples were collected over three stages of pregnancy were included from 2014 to 2016. The urinary concentrations of vanadium were analyzed. Children’s anthropometric parameters were measured at birth, 6, 12 and 24 months. In boys, each doubling increase in vanadium concentrations at middle pregnancy was inversely associated with weight-for-length [− 9.07% (− 17.21%, − 0.93%)] and BMI z-score [− 9.66% (− 18.05%, − 1.28%)] at 24 months. Moreover, vanadium exposure at late pregnancy was negatively associated with weight [− 9.85% (− 16.42%, − 3.28%)], weight-for-length [− 11.00% (− 18.40%, − 3.60%)], and BMI z-scores [− 11.05% (− 18.67%, − 3.42%)] at 24 months in boys. However, the negative associations were not observed in girls, and we found evidence for sex difference (FDR p for interaction = 0.01, 0.01 and 0.03 for weight, weight-for-length and BMI z-scores, respectively). Prenatal vanadium exposure may have an adverse effect on early-childhood growth, and the middle and late pregnancy could be windows of vulnerability for the adverse effects of vanadium exposure on growth development.