Objective: To construct and validate a risk prediction model for sarcopenia in middle-aged and elderly patients with type 2 diabetes mellitus (T2DM) based on inflammation-blood glucose levels. Methods: A retrospective study was conducted on 186 middle-aged and elderly patients with T2DM admitted to Tianjin Beichen hospital from June 2023 to June 2025. Based on the presence or absence of sarcopenia, the patients were categorized into two groups: the sarcopenia group and the non-sarcopenia group.The inflammatory indexes, blood glucose indexes and other data were compared between the two groups. Multivariate Logistic regression was used to analyze the influencing factors of sarcopenia in middle-aged and elderly patients with T2DM. According to the multivariate analysis, a nomogram model was constructed to verify the efficacy of the model. Results: Among 313 patients, 47 (15.02%, 47/313) developed sarcopenia. Multivariate Logistic regression analysis showed that body mass index (OR=0.568, 95%CI=0.339-0.951), HbA1c (OR=2.794, 95%CI=1.251-6.242), IL-6 (OR=1.698, 95%CI=0.339-0.951), and NLR (OR=2.807, 95%CI=1.453-5.423) were the influencing factors for sarcopenia in middle-aged and elderly patients with T2DM (P < 0.05). The calibration curve showed that the model had a good fit, and the decision curve showed that the model had a positive net benefit. The Receiver operating characteristic (ROC) curve showed that the Area under curve (AUC) of the model for predicting sarcopenia in middle-aged and elderly patients with T2DM was 0.967, and the sensitivity was 87.23%. The specificity was 93.61%. The Hosmer-Lemeshow (H-L) goodness of fit test showed that the model had a good fit (χ2=1.621, P=0.982). Conclusion: The inflammation-blood glucose level based risk prediction model has a good predictive effect on the risk of sarcopenia in middle-aged and elderly patients with T2DM, with high sensitivity and specificity.