Objective: To explore the correlation between blind box consumption and non-suicidal self-injury(NSSI) among middle school students, so as to provide new theoretical insights for the prevention of NSSI. Methods: Using stratified random cluster sampling method, 2 807 middle school students aged 11-19 years old were selected from Hunan and Gansu provinces from November 2024 to March 2025. The blind box consumption questionnaire and Functional Assessment of Self mutilation Scale were administered to collect data on students blind box consumption frequency, as well as NSSI behavior. The χ 2 test was used to compare differences in the distribution of NSSI across different groups. Multivariate Logistic regression analysis was performed to infer the correlation and gender differences. Results: A total of 15.3% of middle school students reported having at least one NSSI incident in the past year, among which the reported rates of occasional NSSI (1-4 times) and repeated NSSI (≥5 times) were 5.5% and 9.8% respectively. The results of univariate analysis showed that there was statistically significant different in NSSI distribution among groups with different blind box consumption frequencies ( χ 2=55.72, P <0.05). After adjusting for confounding factors such as age, gender, school stage, family type, discipline style, pocket money, impulsiveness and emotion management, the results of multiple Logistic regression models showed that compared with the group without blind box consumption, the risks of "occasional NSSI" and "repeated NSSI" were higher in the group with blind box consumption ( OR =1.54, 1.66), and the frequency of blind box consumption(continous variable) was positively correlated with the risks of "occasional NSSI" and "repeated NSSI" among middle school students ( OR =1.26, 1.34)(all P <0.05).After gender stratification, the consumption behavior of blind boxes and the frequency of blind box consumption (continuous variable) of boys and girls were associated with "repeated NSSI"(boys: OR =1.61, 1.32, girls: OR =1.65, 1.35), and only in the male group was a correlation between blind box consumption and "occasional NSSI" observed ( OR =2.27) (all P <0.05). Conclusion Blind box consumption may be related to NSSI among middle school students, and there are gender differences in its correlation with NSSI among middle school students.