Arbuscular mycorrhizal fungi (AMF) are a group of ecologically important soil microbes and show wide geographic distribution across the globe. AMF form obligate symbiosis with roots of -80% land plants. In the symbiosis, host plants provide carbon for AMF in return for several benefits, i.e., promoting nutrient uptake, tolerating drought and salt stress, resisting pathogens and herbivores, etc. AMF also can redistribute resources (i.e., C, N and P) between plants and alter their competitive interactions, and thus drive plant population dynamics and community processes. AMF diversity is one of the most important components in soil diversity. In the past decades, AMF are found in almost all terrestrial habitats, including grassland, forest, desert, wetland, alpine meadow, polar region and mangrove, etc. This suggests that AMF have high species diversity. Although AMF diversity has a relatively long research history, most studies only tried to investigate species composition in AMF communities, little is known about the functioning of AMF diversity. In this mini-review, we summarized the new advances in the AMF diversity field, including ecological functioning, determinants and assembling rules. AMF diversity has important ecological functioning. Here, we discussed three aspects: the effects on plant system diversity, stability and productivity. First, several studies reported that AMF diversity is an important determinant for plant diversity. This might be caused by mycorrhizal dependence of subordinate plants. Some studies found that host plants have some preferentially selection towards AMF. Thus, with increasing AMF diversity, subordinate plants will have a higher probability to meet their best AMF partner. Another possibility is that negative plant-mycorrhiza feedbacks might generate positive AMF diversity-plant diversity patterns. This might be caused by host selection towards specific AMF communities. Distinctive AMF communities will make host plants occupy different niche for soil resources. Secondly, AMF diversity could stabilize plant community. Two possibilities can be used to explain this pattern. One is functioning redundancy for several AMF species. In AMF communities with high diversity, loss of certain fungal species will not affect plant community because of similar functioning shared by other AMF. The other possibility is that high AMF diversity will relax competition between different plants for soil nutrients. Third, AMF diversity promotes plant productivity. It seems a general pattern but the mechanisms underling it are still in debate. Complementary effect states that different AMF have different functioning. Higher AMF diversity will have greater functional diversity, which will generate higher plant productivity. Sampling effect states that higher AMF diversity increases the probability of plants encountering the super fungus. Many factors can affect AMF diversity, but here, we are only concentrated in host plants, environmental conditions and anthropogenic disturbance. Host plants affect AMF diversity through different taxonomic levels, including genetype or ecotype, species and community. Environmental conditions include soil nutrient, i.e., N and P, soil type, pH, precipitation and temperature. Anthropogenic disturbance includes agricultural practices, such as tillage, pesticides, fertilization, and land use conversion. Lastly, we discussed the theoretical hypothesis of AMF community assembling process, and provided the analytical methods for dissecting niche and neutral process. At present, it is debating for the assembling process of AMF community in mycorrhizal ecology. In fact, both processes might act simultaneously for AMF community. We deem that future studies should pay attention to the mechanisms underling the positive AMF species diversity-plant productivity relationships, as well as the AMF diversity maintaining mechanisms.