ABSTRACT Previous studies have confirmed a high prevalence of anxiety among medical students. The present study aimed to identify distinct medical student anxiety profiles to develop targeted interventions. A cross-sectional study was conducted with 735 medical students (346 males and 389 females) recruited voluntarily from mainland China. Latent profile analysis (LPA) was applied to explore profiles of various types of anxiety, including test anxiety, academic anxiety, social anxiety, career choice anxiety, general anxiety, and trait anxiety. The results showed that LPA yielded four profiles, comprising 17.14, 20.82, 21.50, and 40.54% of the sample respectively. Specifically, Class 1 (17.14%) showed the highest levels of multiple types of anxiety, particularly the highest level of general anxiety. Class 2 (20.82%) and Class 3 (21.50%) showed moderate levels of anxiety, but they displayed different patterns across various types of anxiety except for social anxiety. Class 4 (40.54%) displayed the lowest levels of anxiety across all types. These profiles differed in academic status and depression. Further, the multinomial logistic regression analysis showed that the high-anxiety group (Class 1) was more likely to be elder students, with lower sports frequency and a higher degree of self-handicapping. These results call for the need to move beyond linear relations among global constructs to address the complexity of anxiety coping and highlight the importance of customized intervention for these heterogeneous groups.