Objective <p>This study aimed to investigate the relationships between physical activity levels, flow state, and depressive symptoms in university students, and to analyze the mediating role of flow state. Methods: A cross-sectional study was conducted, recruiting 3,326 university students from Shanghai Songjiang University Town. Participants were assessed using the Physical Activity Rating Scale-3 (PARS-3), the Flow State Scale (FSS), and the Beck Depression Inventory-II (BDI-II). Statistical analyses, including demographic statistics, preliminary analyses, and Pearson correlation, were performed using SPSS 29.0. Subsequently, Structural Equation Modeling (SEM) was conducted with AMOS 28.0 to test the mediation model, employing 5,000 bootstrap samples to assess the indirect effect. Results: Among the included students, correlation analyses revealed that physical activity level was negatively correlated with depressive symptoms (<i>r</i> = -0.125, <i>p</i> &lt; 0.01) and positively correlated with flow state (<i>r</i> = 0.089, <i>p</i> &lt; 0.01). Flow state was negatively correlated with depressive symptoms (<i>r</i> = -0.229, <i>p</i> &lt; 0.01). The SEM demonstrated a good fit. Mediation analysis indicated a significant total effect of physical activity on depressive symptoms (β = -0.129, 95% CI [-0.163, -0.093], <i>p</i> &lt; 0.001). The direct effect was significant (β = -0.105, 95% CI [-0.140, -0.071], <i>p</i> &lt; 0.001), accounting for 81.4% of the total effect. The indirect effect through flow state was also significant (β = -0.024, 95% CI [-0.036, -0.013], <i>p</i> &lt; 0.001), accounting for 18.6% of the total effect. Conclusion: Physical activity exerts both direct and indirect alleviating effects on depressive symptoms among university students, with flow state playing a significant partial mediating role. Intervention strategies should emphasize enhancing physical activity and optimizing exercise design to foster flow experiences.</p>