BACKGROUND: Differentiating psychotic major depression (PMD) from non-PMD (NPMD) is crucial as it influences treatment decisions, prognosis, and patient outcomes. This study aims to develop an efficient model for precision diagnostics of PMD based on clinical indicators. METHODS: A total of 731 patients who visited our hospital with major depression (MD) were enrolled, including a discovery cohort and a validation cohort. We retrospectively analyzed the distribution differences of 20 clinical indicators in the discovery cohort. We included difference clinical indicators (DCIs) in the logistic regression algorithm analysis to establish a multi-panel detection model. The model's performance in distinguishing PMD from NPMD and in distinguishing bipolar MD from MD was validated in the validation cohort by the receiver operator characteristic curve (ROC), the area under the curve (AUC), sensitivity, and specificity. RESULTS: We have constructed a six-DCIs diagnosis model (6MP) based on the discovery cohort. The AUC of 6MP for identifying PMD and NPMD was 0.826, and the sensitivity and specificity were 87.5 % and 70.59 %, respectively. In the validation cohort, the AUC for identifying PMD and NPMD was 0.781, and the sensitivity and specificity were 78.99 % and 67.31 %. The AUC for identifying bipolar MD and MD without psychotic symptoms was 0.716, and the sensitivity and specificity were 60.71 % and 76.92 %. CONCLUSIONS: This study not only provides new tools for the precise diagnosis and treatment of PMD but also hopes to simplify the diagnostic process, improve the specificity of treatment, and thus bring more practical clinical benefits to patients.