Polycystic ovary syndrome (PCOS) is an endocrine disease with metabolic disorders, and patients with PCOS often have impaired reproductive potential. Examining the composition of follicular fluid (FF) in these patients can help identify the specific changes that may impact oocyte quality and potentially cause infertility. This study sought to examine the differential variations in the metabolic composition of FF among normal-weight patients with PCOS and controls and to identify potential biomarkers that may offer insights into the early identification and management of these patients. We collected FF samples from 45 normal-weight women with PCOS and 36 normal-weight controls without PCOS who were undergoing in vitro fertilization-embryo transfer. Untargeted metabolomic study of collected FF from infertile women was performed using high-performance liquid chromatography–tandem spectrometry (LC-MS). The tendency of the two groups to separate was demonstrated through multivariate analysis. Univariate analysis and variable importance in the projection were used to screen out differential metabolites. Metabolic pathway analysis was conducted using the Kyoto Encyclopedia of Genes and Genomes (KEGG), and a diagnostic model was established using the random forest algorithm.