<p>Understanding the genesis of galena in Zn-Pb deposits is critical for gaining insights into ore formation processes, which are influenced by factors such as temperature, fluid composition, and metal sources. While conventional approaches often rely on trace element compositions to classify deposit types, distinguishing formation environments remains challenging due to overlapping geochemical signatures. This study demonstrates the capacity of machine learning (ML) to classify the genesis of galena in the Zawar Zn-Pb deposit, Rajasthan, India, using in-situ Laser Ablation-Inductively Coupled Plasma Mass Spectrometry trace element data. We framed this as a multi-classification problem, using Random Forest and Gradient Boosting algorithms to process complex, non-linear datasets. A comprehensive dataset of trace element compositions from 975 published galena samples across globally distributed Epithermal, Mississippi Valley Type (MVT), SEDEX, CRD, and Vein-type deposits was used to train our models. The model performance was then evaluated on the test set, achieving overall accuracies of ~ 97.51% with Random Forest and ~ 96.37% with Gradient Boosting. Finally, we applied these trained models to trace element data from Zawar galena samples, suggesting a MVT-type hydrothermal origin for the Zawar deposit. These results highlight the potential of machine learning in addressing complex geological challenges, such as deposit origin identification.</p>