Background <p>Lung immune prognostic index (LIPI) is associated with survival outcomes in patients with non-small cell lung cancer (NSCLC) receiving immunotherapy, but the association with the occurrence of checkpoint inhibitor pneumonitis (CIP) is unclear.</p> Methods <p>We retrospectively included 1824 patients with advanced NSCLC who received immune checkpoint inhibitors (ICIs) at two institutions. Cox regression analysis and cumulative incidence curve were used to evaluate the predictive value of LIPI. Additionally, we performed competing risk analysis using Fine-Gray regression and cumulative incidence curves.</p> Results <p>During a median follow-up of 15 months, 99 patients developed CIP. Compared with the good LIPI group, the intermediate LIPI group (HR 1.87, <i>P</i> = 0.007) and poor LIPI group (HR 4.39, <i>P</i> &lt; 0.001) had a higher risk of CIP. Furthermore, we found LIPI was an independent predictor for intermediate-grade CIP (intermediate LIPI: HR 2.11, <i>P</i> = 0.056; poor LIPI: HR 4.51, <i>P</i> = 0.002) and high-grade CIP (intermediate LIPI: HR 6.94, <i>P</i> = 0.014; poor LIPI: HR 44.01, <i>P</i> &lt; 0.001), but not for low-grade CIP (intermediate LIPI: HR 1.31, <i>P</i> = 0.392; poor LIPI: HR 0.72, <i>P</i> = 0.656). Similar results were obtained after competing risk analysis.</p> Conclusions <p>LIPI grade shows potential in predicting the risk of CIP during immunotherapy and could be valuable in clinical management.</p>