Aboveground herbivores have strong effects on grassland nitrogen (N) cycling. They can accelerate or slow down soil net N mineralization depending on ecosystem productivity and grazing intensity. Yet, most studies only consider either ungulates or invertebrate herbivores, but not the combined effect of several functionally different vertebrate and invertebrate herbivore species or guilds. We assessed how a diverse herbivore community affects net N mineralization in subalpine grasslands. By using size‐selective fences, we progressively excluded large, medium, and small mammals, as well as invertebrates from two vegetation types, and assessed how the exclosure types (ET) affected net N mineralization. The two vegetation types differed in long‐term management (centuries), forage quality, and grazing history and intensity. To gain a more mechanistic understanding of how herbivores affect net N mineralization, we linked mineralization to soil abiotic (temperature; moisture; NO3−, NH4+, and total inorganic N concentrations/pools; C, N, P concentrations; pH; bulk density), soil biotic (microbial biomass; abundance of collembolans, mites, and nematodes) and plant (shoot and root biomass; consumption; plant C, N, and fiber content; plant N pool) properties.Net N mineralization differed between ET, but not between vegetation types. Thus, short‐term changes in herbivore community composition and, therefore, in grazing intensity had a stronger effect on net N mineralization than long‐term management and grazing history. We found highest N mineralization values when only invertebrates were present, suggesting that mammals had a negative effect on net N mineralization. Of the variables included in our analyses, only mite abundance and aboveground plant biomass explained variation in net N mineralization among ET. Abundances of both mites and leaf‐sucking invertebrates were positively correlated with aboveground plant biomass, and biomass increased with progressive exclusion. The negative impact of mammals on net N mineralization may be related partially to (1) differences in the amount of plant material (litter) returned to the belowground subsystem, which induced a positive bottom‐up effect on mite abundance, and (2) alterations in the amount and/or distribution of dung, urine, and food waste. Thus, our results clearly show that short‐term alterations of the aboveground herbivore community can strongly impact nutrient cycling within ecosystems independent of long‐term management and grazing history.