Background: Several studies to date have reported on the development of positron emission tomography (PET)/computed tomography (CT)-based models intended to effectively distinguish between benign and malignant pulmonary nodules (PNs). This meta-analysis was designed with the goal of clarifying the utility of these PET/CT-based conventional parameter models as diagnostic tools in the context of the differential diagnosis of PNs. Methods: Relevant studies published through September 2023 were identified by searching the Web of Science, PubMed, and Wanfang databases, after which Stata v 12.0 was used to conduct pooled analyses of the resultant data. Results: This meta-analysis included a total of 13 retrospective studies that analyzed 1,731 and 693 malignant and benign PNs, respectively. The respective pooled sensitivity, specificity, PLR, and NLR values for the PET/CT-based studies developed in these models were 88% (95%CI: 0.86–0.91), 78% (95%CI: 0.71–0.85), 4.10 (95%CI: 2.98–5.64), and 0.15 (95%CI: 0.12–0.19). Of these endpoints, the pooled analyses of model sensitivity (I2 = 69.25%), specificity (I2 = 78.44%), PLR (I2 = 71.42%), and NLR (I2 = 67.18%) were all subject to significant heterogeneity. The overall area under the curve value (AUC) value for these models was 0.91 (95%CI: 0.88–0.93). When differential diagnosis was instead performed based on PET results only, the corresponding pooled sensitivity, specificity, PLR, and NLR values were 92% (95%CI: 0.85–0.96), 51% (95%CI: 0.37–0.66), 1.89 (95%CI: 1.36–2.62), and 0.16 (95%CI: 0.07–0.35), with all four being subject to significant heterogeneity (I2 = 88.08%, 82.63%, 80.19%, and 86.38%). The AUC for these pooled analyses was 0.82 (95%CI: 0.79–0.85). Conclusions: These results suggest that PET/CT-based models may offer diagnostic performance superior to that of PET results alone when distinguishing between benign and malignant PNs. [ABSTRACT FROM AUTHOR]