Objective To systematically evaluate and analyze the risk prediction models for pneumonia in patients with traumatic brain injury （TBI）, so as to provide references for clinical practice. Methods We searched Pubmed, Web of science, Cochrane library, EMBASE, Wanfang database, CNKI, VIP database, CBM to collect studies published during the period from database inception to August 2nd, 2021, on risk prediction models for pneumonia in TBI patients. Two researchers independently screened the literature，extracted information，and assessed the risk of bias and applicability of the included literature by using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Results Six relevant studies totaling 6 risk prediction models for pneumonia in TBI patients were included. The area under the ROC curve of the models ranged from 0.806 to 0.949. The most commonly reported predictive factors in all included models encompassed GCS score, mechanical ventilation, post-operative albumin level, and APACHE Ⅱscore. All included studies were subjected to a certain level of bias, and the applicability was unclear, which was mostly due to such shortcomings as not reporting blinding method, insufficient sample size, overfitting of model , not reporting or not handling missing data, and lack of model performance assessment. Conclusion The research on risk prediction models for pneumonia in TBI patients is still in the development stage. In the future, researchers could conduct multi-center and large sample research, and in conjunction with big data processing technique, develop easy-to-use risk prediction models with excellent performance. Researchers should continue to update and correct the models during use, in an effort to provide a solid predictive model for clinical settings.