Abstract: The West Qinling orogen is an important gold and polymetallic metallogenic belt in China, where orogenic, Carlin and Carlin-like gold deposits are widely distributed. Nevertheless, there remain considerable controversies over the genetic classification of gold deposits based on conventional geological studies. Pyrite is a dominant gold-bearing mineral in gold deposits, and its trace element compositions can effectively constrain metallogenic conditions and deposit genesis. To achieve efficient and accurate discrimination of genetic types of gold deposits in the West Qinling orogen, this study systematically compiled 3062 sets of pyrite trace element data from 15 typical gold deposits in the region. Combined with machine learning algorithms including Support Vector Machine (SVM) and Random Forest (RF), as well as multivariate statistical methods such as Principal Component Analysis (PCA), classification models were established. The classification accuracies of the SVM and RF models reach 94.2% and 97.1%, respectively. The results reveal that Au, As and Sb in pyrite are the key indicator elements for distinguishing the three types of gold deposits. On this basis, a ternary discrimination diagram of ln(Au)–ln(As)–ln(Sb) for pyrite was constructed, which can effectively differentiate the three genetic types of gold deposits. This study provides a new approach for genetic type discrimination of gold deposits in the West Qinling orogen, and also offers a reference case for the application of machine learning in mineral deposit research.