Objective: This meta-analysis aimed to evaluate the comparative diagnostic performance of amyloid-beta positron emission tomography (A beta PET) and magnetic resonance imaging (MRI) in diagnosing Alzheimer's disease (AD). Methods: An extensive search was conducted in the PubMed and Embase databases to identify available publications up to December 2023. Head-to-head comparative studies were included if they evaluated the diagnostic performance of A beta PET and MRI in diagnosing Alzheimer's disease. Sensitivity and specificity were assessed using the DerSimonian and Laird method, followed by transformation via the Freeman-Tukey double inverse sine transformation. Results: Six articles involving 560 patients were included in the meta-analysis. When distinguishing AD from mild cognitive impairment (MCI), both methods showed comparable sensitivity (A beta PET: 0.71, MRI: 0.62) and specificity (A beta PET: 0.68, MRI: 0.69), with no statistically significant differences observed (p = 0.34 and 0.99). When identifying AD from normal cognitive control (NC), both A beta PET and MRI showed similar results, with comparable sensitivity (A beta PET: 0.93, MRI: 0.85) and specificity (A beta PET: 0.95, MRI: 0.82), without significant differences (p = 0.38 and 0.19). Similarly, in detecting MCI from NC, both A beta PET and MRI demonstrated similar sensitivity (A beta PET: 0.69, MRI: 0.64) and specificity (A beta PET: 0.75, MRI: 0.76) without significant differences (p = 0.40 and 0.94). However, 18F-FMM seems to have a higher specificity compared to MRI when distinguishing AD from MCI (P = 0.03) and AD from NC (p = 0.04). Conclusions: Our meta-analysis indicates that A beta PET demonstrates similar sensitivity and specificity to MRI in diagnosing Alzheimer's disease. However, the limited number of studies may impact the evidence of the current study; further larger sample prospective research is required to confirm these findings.