Recurrent implantation failure (RIF) refers to repeated failure to become pregnant after transferring embryos with normal morphology. However, the pathogenesis of RIF remains unrevealed, especially for those without any pathological features. In this study, we characterized the vaginal microbiota and metabolomes of patients with unexplained RIF, while patients who achieved clinical pregnancy in the first frozen embryo transfer (FET) cycle were used as controls. Based on 16S rRNA gene sequencing of the vaginal microbiota, the vaginal showed a significant positive correlation with the pregnancy rate, and the RIF group presented higher microbial α-diversity than the control group ( value = 0.016). The metabolomic profile identified 2,507 metabolites, of which 37 were significantly different between the two groups ( value < 0.05, variable importance for the projection [VIP] > 1). Among them, 2',3-cyclic UMP and inositol phosphate were the top two metabolites that were higher in the RIF group, while glycerophospholipids and benzopyran were important metabolites that were lower in the RIF group. A lack of lysobisphosphatidic acid and prostaglandin metabolized from glycerophospholipids will lead to deferred implantation and embryo crowding. Benzopyran, as a selective estrogen receptor modulator, may affect the outcome of pregnancy. All of the changes in metabolite profiles may result in or from the differential microbiota compositions in RIF patients. In conclusion, significant differences were presented in the vaginal microbiota and metabolomes between patients with unexplained RIF and women who became pregnant in the first FET cycle. For the first time, this study elaborates the possible pathogenesis of RIF by investigating the vaginal microbiota and metabolites in RIF patients. fertilization-embryo transfer (IVF-ET) is now widely applied for treating infertility, and unexplained recurrent implantation failure (RIF) has become a substantial challenge. We hypothesize that vaginal microbial dysbiosis is associated with RIF, as it is linked to many female reproductive diseases. In this study, we characterized the vaginal microbiota and metabolomes of patients with unexplained RIF, while patients who achieved clinical pregnancy in the first IVF cycle were set as controls. In general, significant differences were discovered in the vaginal microbiota and metabolomes between the two groups. This study is the first detailed elaboration of the vaginal microbiota and metabolites associated with RIF. We believe that our findings will inspire researchers to consider the dynamics of microbiomes related to the microenvironment as a critical feature for future studies of nosogenesis not only for RIF but also for other reproductive diseases.