Multiple myeloma (MM) is the second most commonly diagnosed hematological malignancy. Understanding the basic mechanisms of the metabolism in MM may lead to new therapies that benefit patients. We collected the gene expression profile data of GSE39754 and performed differential analysis. Furthermore, identify the candidate genes that affect the prognosis of the differentially expressed genes (DEGs) related to the metabolism. Enrichment analysis is used to identify the biological effects of candidate genes. Perform coexpression analysis on the verified DEGs. In addition, the candidate genes are used to cluster MM into different subtypes through consistent clustering. Use LASSO regression analysis to identify key genes, and use Cox regression analysis to evaluate the prognostic effects of key genes. Evaluation of immune cell infiltration in MM is by CIBERSORT. We identified 2821 DEGs, of which 348 genes were metabolic-related prognostic genes and were considered candidate genes. Enrichment analysis revealed that the candidate genes are mainly related to the proteasome, purine metabolism, and cysteine and methionine metabolism signaling pathways. According to the consensus clustering method, we identified the two subtypes of group 1 and group 2 that affect the prognosis of MM patients. Using the LASSO model, we have identified 10 key genes. The prognosis of the high-risk group identified by Cox regression analysis is worse than that of the low-risk group. Among them, PKLR has a greater impact on the prognosis of MM, and the prognosis of MM patients is poor when the expression is high. In addition, the level of immune cell infiltration in the high-risk group is higher than that in the low-risk group. In the summary, metabolism-related genes significantly affect the prognosis of MM patients through the metabolic process of MM patients. PKLR may be a prognostic risk factor for MM patients.