Abstract: Organic agriculture is conducive to increasing plant diversity in farmland, and rich plant communities can provide animals with a range of services, including sources of pollen for honey, host, and habitats. Although there have been many studies of the effect of organic management on plant α diversity in single farmland habitat, little research into β and γ diversity has been conducted. In this study, a large area of organically managed farmland and a nearby conventionally managed farmland with multi-habitats were selected, and the differences in the α, β, and γ diversity of herbaceous plants between two management practices and among different habitat types were studied. In both cases, the farmland habitats included vegetable greenhouse, orchard, open-air field, field boundary, and paddy ridge. The herb plant survey was conducted at the start of summer. Findings from ANOVA revealed that, though the average number of local herb plants and medicinal herb plants (α diversity) was significantly higher in all habitats in organic farmland than the case in conventional farmland, organic management only increased the number of herb plant and medicinal herb plant species in open-air field and the number of medicinal herb plant species in paddy ridge. The impact of organic management on other habitats was not significant. Under conventional management, there were no significant differences in the number of plant species among five habitat types. However, under organic management, the number of herb plant species in open-air field and medicinal herb plant species in paddy ridge were significantly higher than those in vegetable greenhouse. Results of non-metric multi-dimensional analysis based on the Manhattan index indicated that, although there was no difference in the overall species composition (β diversity) of organically managed farmland and conventionally managed farmland, there were significant differences between organic management and conventional management in the herb plant and medicinal herb plant species compositions of open-air field and paddy ridge. β diversity was higher in organic farmland than in conventional farmland, and relatively independent plant communities formed in different habitats. A rarefaction curve of species richness indicated that, although the total number of herb plant species and the number of medicinal herb species (γ diversity) in organically managed farmland were higher than those in conventionally managed farmland, the difference was not significant statistically. Organic management increased the α diversity and β diversity of herbaceous plants in both paddy ridge and open-air field, but there was no obviously beneficial effect of organic management in vegetable greenhouse, field boundary, and orchard. The increase in local α diversity and β diversity did not necessarily result in a significant increase in γ diversity. In summary, it cannot be concluded universally that organic agriculture can increase plant diversity in different habitats and at different spatial scales. It is necessary to specifically consider the differences in different habitats and diversity indices at different spatial scales. Longer implementation of organic management, low-frequency weeding, and artificial increases in the pool of local plant species can help increase herbaceous plant diversity of the entire region and in all habitat types.