Abstract: The Qimantagh Orogenic Belt in the East Kunlun is an important Cu−Mo−Fe−Pb−Zn polymetallic mineralization belt in the northwest of China, and many porphyry-skarn deposits that are genetically related to granitoids are founded, such as Kaerqueka, Yemaquan, Weibao, and Wulanwuzhuer. With the development of a new round of strategic action to find mineral breakthroughs, further strengthening the study of granite mineralization potential in the Qimantagh Orogenic Belt has become an important breakthrough to promote the growth of metal mineral reserves in the region. In this paper, based on the systematic collection of whole−rock major and trace element data of mineralized and barren magmatic rocks of typical porphyry−skarn polymetallic deposits in the Qimantagh Orogenic Belt, 28 common whole-rock geochemical features are selected, and the machine learning algorithm (Random Forest) is used for the training of the machine learning model to establish a machine learning model capable of identifying the mineralized and barren magmatic rocks of porphyry−skarn polymetallic deposits in the region. A new method is developed to identify the mineralized and barren magmatic rocks in the porphyry−skarn polymetallic deposits in this area. According to the model evaluation metric, the accuracy of the Random Forest classification model trained in this paper is 0.90, which proves that the method can effectively recognize mineralized and barren magmatic rocks. This study provides a new idea for the prospecting and exploration of porphyry−skarn polymetallic deposits in the Qimantagh Orogenic Belt, which will greatly improve the efficiency of prospecting, reduce the economic and labor costs of prospecting, and thus better serve the new round of strategic action of prospecting and breakthrough. The machine learning code has been uploaded to GitHub at https://github.com/ShihuaZhong/2023-Qimantagh-RF-whole-rock-classifier.