This study, conducted in Guizhou Province, China, utilized environmental DNA (eDNA) technology to conduct a fish diversity survey in 30 typical mountainous rivers across seven river systems. The results of the study revealed a total of 89 fish species, with 83 of these species being identified to the species level. The majority of the identified species belonged to the order Cypriniformes, including five exotic species and four rare or endangered species. It is noteworthy that at least 74 species (89.2% of those identified to the species level) have been previously documented in existing studies .With respect to their habitat, benthic and demersal fish were predominant, while omnivorous and carnivorous species were more prevalent based on their feeding habits.Cyprinus carpio, Carassius auratus, Hypophthalmichthys nobilisand Rhinogobius giurinu were found to be widely distributed, appearing in over 90% of the sampling sites. Fish diversity indices (Richness, Shannon-Wiener, Simpson, and Pielou) indicated favorable fish diversity conditions in the surveyed rivers, with no significant altitudinal gradient patterns observed in the diversity indices. Utilizing piecewise structural equation modelling, we conducted a comprehensive analysis of the factors influencing the four fish diversity indices. These factors included altitude, landscape type, physicochemical parameters, pollution indicators, ecological flow satisfaction, and human activities. The results of the analysis revealed that physicochemical parameters and pollution indicators emerged as the primary factors affecting fish diversity in the mountainous rivers of Guizhou Province. In contrast, human activities and ecological flow satisfaction were found to be secondary factors. Furthermore, an aggregated boosted treeanalysis was employed to assess the importance of various water quality parameters on fish diversity indices, revealing that water temperature, conductivity, and ammonia nitrogen were the primary water quality factors affecting fish diversity. Additionally, among the environmental factors examined, pollution indicators, human activities, and ecological flow satisfaction directly influenced fish diversity indices, whereas other factors exerted indirect effects. Notably, different diversity indices exhibited varying responses to the same environmental factors. This study provides methodological insights for fish diversity surveys in mountainous rivers and offers a data-driven foundation and theoretical support for the conservation of fish biodiversity in these ecosystems.