Background: Violence-exposed youth rarely receive mental health services, even though exposure increases risk for academic and psychosocial problems. This study examines the association between violence exposure and mental health service contact. The 4 forms of violence exposure were peer, family, sexual, and witnessing. Methods: Data are from 1534 Boston public high school students who participated in a 2008 self-report survey of violence exposure and its correlates. Multivariate logistic regressions estimated associations between each form of violence with service contact, then examined whether associations persisted when controlling for suicidality and self-injurious behaviors. Results: In unadjusted models, violence-exposed students more often reported service contact than their peers. However, in multivariate models, only exposure to family (odds ratio [OR] = 1.69, 95% confidence interval [CI] = 1.23-2.31) and sexual violence (OR = 2.34, 95% CI = 1.29-4.20) were associated with service contact. Associations attenuated when controlling for suicidality and self-injurious behaviors, indicating they were largely explained by self-harm. Sexual violence alone remained associated with mental health service contact in fully adjusted models, but only for girls (OR=3.32, 95% CI=1.30-8.45), suggesting sex-specific pathways. Conclusions: Associations between adolescent violence exposure and mental health service contact vary by forms of exposure. Outreach to a broader set of exposed youth may reduce the impact of violence and its consequences for vulnerable students.