Background. The pretreatment albumin and globulin ratio (AGR) was an inflammation-associated factor which was related to the overall survival in various malignancies. The aim of this study was to evaluate the prognostic value of AGR in patients with gastric cancer. Method. This retrospective study included 862 cases pathologically diagnosed with gastric cancer. All patients were randomly divided into the testing group (431 cases) and validation group (431 cases). The relationships of AGR with clinicopathologic characteristics and prognosis were analyzed by Kaplan-Meier and Cox regression methods. Results. In the testing group, the median overall survival was 26.90 months and the cutoff value of AGR was 1.50 based on R language. Kaplan-Meier analysis showed that lower AGR was correlated with poorer overall survival. Multivariate analysis demonstrated that AGR was an independent prognostic factor for overall survival (HR: 0.584, 95% CI = 0.351-0.973, and p = 0.039). In the validation group, the median overall survival was 24.10 months. Lower AGR (≤1.50) also had a significantly poorer overall survival by Kaplan-Meier analysis. According to multivariate analysis, the AGR was also confirmed to be an independent prognostic factor for overall survival (HR: 0.578, 95% CI = 0.373-0.897, and p = 0.015). Conclusions. Our study suggested that the pretreatment AGR could be a prognostic biomarker for overall survival in patients with gastric cancer.