In the context of worldwide biodiversity and wild bee decline, it is increasingly important to better understand the effect of land-use changes on wild bee communities at a global scale. To do so, we studied the effect of city area and urban green spaces layout on wild bee species richness and community composition, as well as on wild bee species with an unfavorable UICN conservation status. This study was based on a large European dataset encompassing 20 cities from France, Belgium and Switzerland. We found a mean wild bee species richness in cities of 96 ± 48 (SD), showing that this species richness was highly variable among cities. The main factor positively influencing wild bee species richness in cities was the area of the city. Conversely, species richness was not significantly related to the total area of urban green spaces in a given city, measured as the spatial extent of urban parks, wastelands and other semi-natural habitats, excluding urban private gardens. Species with conservation status were quite scarce in urban environments, especially when compared to the European Red List of Bees, and we could not link their presence to either city or urban green space area. Dissimilarities in wild bee species community compositions were not associated with any of the studied characteristics of cities. We found that the dissimilarity of wild bee community composition among cities was mainly driven by the rarest species, as the most common ones were found in a majority of the cities sampled. Overall, these results emphasize that larger cities host more wild bee species, but are no refuge for the ones with concerning conservation status. Thus, stakeholders are encouraged to design their cities in favor of biodiversity to better support wild bee communities, and perhaps mitigate the established effect of the urban ecological filter.