Background and Aim: The prevalence of lean/nonobese nonalcoholic fatty liver disease (NAFLD) ranges widely in studies. Thus, here, we aimed to perform a meta-analysis on NAFLD prevalence in the lean or nonobese population to give clarity. Materials and Methods: PubMed, Embase, and the Cochrane Library databases were systematically searched to identify studies reporting NAFLD prevalence in the lean/nonobese population. Lean or nonobese was defined by body mass index cutoffs reported by authors in original studies. NAFLD prevalence based on community, population, or health checkups was combined with random-effect model after logit transformation. Subgroup analysis and meta-regression were further performed to investigate the heterogenicity. Results: A total of 45 studies were enrolled in the final analysis, with 55,936 lean/nonobese subjects included, among whom 7351 NAFLD patients were diagnosed. Overall, the pooled NAFLD prevalence of the lean or nonobese population was 10.2% (95% confidence interval: 7.6%-13.6%) and 15.7% (95% confidence interval: 12.5%-19.6%), respectively. Compared with western studies, the NAFLD prevalence in the lean or nonobese population was lower in eastern studies. In addition, the NAFLD prevalence in both the lean and nonobese population showed a general upward trend during recent years. The prevalence was similar in community-based and health checkup-based studies. Lean/nonobese NAFLD patients had significantly lower rates of hypertension, lower uric acid and fasting plasma glucose, and a higher level of high-density lipoprotein than nonlean/obese patients. Conclusions: The prevalence of NAFLD in the lean/nonobese population is not rare in either the western or eastern regions of the world. This meta-analysis of prevalence assessment and clinical characteristics should enable higher confidence in more specific interventions and health care standards for these patients.