Infection, a significant cause of death in multiple myeloma (MM) patients, is a common complication and is closely associated with immunoparesis. There exists no clear definition of early infection, so early infection is defined in this paper as the occurrence within 3 months after diagnosis, considering the high incidence of infections within 3 months after diagnosis. This study established a new nomogram model based on immunoparesis to identify MM patients with high-risk early infection. Methods: A retrospective collection of 430 NDMM patients from June 2013 to June 2022 was conducted, and the patients were further divided into a training cohort and a validation cohort. In the training cohort, the least absolute shrinkage and selection operator (LASSO) was used to select the best variables that can be used to establish a new nomogram prediction model. Validation was performed in the validation and entire cohorts. Results: After diagnosis, 67.7 % of the patients suffered from severe infection within 1 year, and 59.5 % experienced the first severe infection within 3 months. Variables associated with an increased risk of severe infection in the first 3 months included: BMPC, D-dimer, serum β2 microglobulin, immunoparesis, albumin, and eGFR. The nomogram based on the above six factors achieved a good C-index of 0.754, 0.73, and 0.731 in predicting early infection in the training cohort, validation cohort, and entire cohort, respectively. Finally, the time-dependent receiver operating characteristic (ROC) curve and decision curve analysis (DCA) of the nomogram showed that the model provided superior diagnostic capacity and clinical net benefit. Conclusion: In this study, we established a nomogram model to predict early grade ≥ 3 infection in NDMM patients. This model can assist clinicians in identifying NDMM patients with high-risk infections and improve their prognosis through early intervention.