Objective To investigate the levels and clinical significance of serum microRNA-625 (miR-625) and serum microRNA-203 (miR-203) in patients with multiple myeloma (MM). Methods A total of 103 patients diagnosed as MM in the hospital from June 2021 to August 2023 were selected as observation group, and they were classified into stage Ⅰ (n=23), stage Ⅱ (n=31) and stage Ⅲ (n=49) according to the International Staging System (ISS) for multiple myeloma. The patients were divided into newly diagnosed group (n=75) and recurrent group (n=28) based on their diagnosis and treatment status. Additionally, 103 healthy individuals with physical examination in the hospital in the same period were selected as control group. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) was used to determine the levels of serum miR-625 and miR-203. A multivariate Logistic regression model was used to analyze the influencing factors of MM. The diagnostic value of serum miR-625 and miR-203 for MM patients was evaluated by receiver operating characteristic (ROC) curve. Results The levels of protein M, hemoglobin, albumin, miR-625 and miR-203 in the observation group were significantly lower than those in the control group, while the levels of serum creatinine, 24-hour urine protein, urea nitrogen, serum β₂-microglobulin and serum calcium were significantly higher than those in the control group (P < 0.05). Serum levels of miR-625 and miR-203 in newly diagnosed MM group were significantly higher than those in the recurrent group (P < 0.05). The serum levels of miR-625 and miR-203 in patients with stage Ⅱ and Ⅲ were significantly lower than those in patients with stageⅠ, and the levels of miR-625 and miR-203 in patients with stage Ⅲ were significantly lower than those in patients with stage Ⅱ (P < 0.05). The serum miR-203 level in MM patients with osteolytic lesions less than 2 was significantly lower than that in patients with osteolytic lesions greater than or equal to 2 (P < 0.05). Serum miR-625 level in MM patients with grades 1 to 2 of bone disease was significantly higher than that in patients with grades 3 to 4 of bone disease (P < 0.05). The miR-625 and miR-203 were protective factors for the occurrence of MM (P < 0.05). The area under the curve (AUC) of serum miR-625 and miR-203 for diagnosing MM was 0.808 and 0.866 respectively, and the AUC of the combineddiagnosis of miR-625 and miR-203 was 0.919. The combined diagnosis of miR-625 and miR-203 was superior to the individual diagnosis of serum miR-625 and miR-203 (Z_{combined-miR-625}=3.816, Z_{combined-miR-203}=2.157, P=0.001, 0.031). Conclusion Serum levels of miR-625 and miR-203 decrease in MM patients, and their combination has a high diagnostic value for MM.