Objective To analyze the risk factors for postoperative intra-abdominal infection in gastric cancer patients, as well as construct and validate a nomogram prediction model. Methods 588 gastric cancer surgery patients who admitted to the Department of General Surgery of the First Affiliated Hospital of Soochow University from April 2021 to March 2024 were selected as the study subjects. Clinical data of patients were collected and randomly divided into the training set and the validation set according to the ratio of 3 ∶1. Clinical data between two groups of patients were compared. Patients were divided into the infection group and non-infection group according to whether they had intra-abdominal infection after surgery. Univariate and multivariate analyses were conducted, and a nomogram prediction model was constructed and validated based on the results of multivariate analysis. Results Among the 588 patients, 52 (8.84%) had postoperative intra-abdominal infection. A total of 65 strains of pathogens were detected from 52 peritoneal fluid specimens, out of which 47 (72.31%) were Gram-negative bacteria, 15 (23.07%) were Gram-positive bacteria, and 3 (4.62%) were fungi. Multivariate logistic regression analysis showed that the degree of eradication (microscopic residue), combined organ resection, hypertension, history of abdominal surgery, and duration of surgery were all independent risk factors for postoperative intra-abdominal infection in gastric cancer patients (all P<0.05). Based on multivariate analysis results, a nomogram prediction model for postoperative intra-abdominal infection in gastric cancer patients was constructed. The receiver operating characteristic (ROC) curve result showed that the areas under the ROC curve (AUCs) of the training set and validation set were 0.764 (95%CI: 0.677-0.852) and 0.712 (95%CI: 0.565-0.860), respectively, indicating that the model had good discriminability for postoperative intra-abdominal infection in gastric cancer patients. Hosmer-Lemeshow test showed a χ² value of 8.491 and a P value of 0.387, suggesting goodness fit of the model. The decision curve analysis (DCA) result showed that within the risk threshold ranges of the training set (0.05-0.4) and validation set (0.1-1.0), positive benefits may be obtained by using the model to intervene in patients with high risk of postoperative intra-abdominal infection. Clinical impact curve (CIC) analysis result showed that within the risk threshold ranges of the training set (0-0.4) and validation set (0-0.5), the number of infected cases predicted by the model was higher than the actual number, indicating good clinical practicality of the model. Conclusion Construction of a nomogram prediction model based on independent risk factors for postoperative intra-abdominal infection in gastric cancer can provide a quantitative and intuitive reference for the early clinical assessment of postoperative intra-abdominal infection in gastric cancer.