Introduction: Neuroendocrine tumors (NETs) are a heterogeneous group of epithelial tumors originating from different anatomical sites, and identifying the gut microbiota and metabolic mechanisms involved in the onset of NETs may help to develop appropriate disease prevention and monitoring strategies. Methods: We employed a mediated two-sample Mendelian randomization (MR) approach, analyzing gut microbiota from German studies and NET datasets from the 10th round of the FinnGen project. Mediation analyses were conducted using the metabolites dataset from the Canadian Longitudinal Study of Aging (CLSA) and the TwinsUK study. Instrumental variables were chosen according to established MR criteria and analyzed using the Wald ratio, inverse-variance weighted (IVW), MR-Egger, and weighted median methods. To ensure robustness, sensitivity analyses were performed using Cochrane’s Q, Egger’s intercept, MR-PRESSO, and leave-one-out methods. Results: Causal relationships were identified between the genetic determinants of 6, 5, 2, 1, 2, 3 gut microbiotas and the risk of colorectal, lung, pancreatic, rectum, small intestine, and stomach NETs. Similarly, the genetic determinants of 4, 6, 1, 5, 10, and 7 metabolites were found to be causally related to the risk of colorectal, lung, pancreatic, rectum, small intestine, and stomach NETs, respectively. Through Wald ratio and IVW methods, we preliminarily identified 957 microbiota-metabolite pairs with significant causal associations and formed 13 mediated relationships between the impact of gut microbiotas on NETs. Conclusion: Our study suggests that gut microbiotas and its derived metabolites may contribute to the onset of NET, offering a novel insight into the disease’s pathogenesis.