Identification of ecosystem services, i.e. the contributions that ecosystems make to human well-being, has proven instrumental in galvanising public and political support for safeguarding biodiversity and its benefits to people. Here we synthesise the global evidence on ecosystem services provided and disrupted by freshwater bivalves, a heterogenous group of >1200 species, including some of the most threatened (in Unionida) and invasive (e.g. Dreissena polymorpha) taxa globally. Our systematic literature review resulted in a data set of 904 records from 69 countries relating to 24 classes of provisioning (N = 189), cultural (N = 491) and regulating (N = 224) services following the Common International Classification of Ecosystem Services (CICES). Prominent ecosystem services included (i) the provisioning of food, materials and medicinal products, (ii) knowledge acquisition (e.g. on water quality, past environments and historical societies), ornamental and other cultural contributions, and (iii) the filtration, sequestration, storage and/or transformation of biological and physico-chemical water properties. About 9% of records provided evidence for the disruption rather than provision of ecosystem services. Synergies and trade-offs of ecosystem services were observed. For instance, water filtration by freshwater bivalves can be beneficial for the cultural service 'biomonitoring', while negatively or positively affecting food consumption or human recreation. Our evidence base spanned a total of 91 genera and 191 species, dominated by Unionida (55% of records, 76% of species), Veneroida (21 and 9%, respectively; mainly Corbicula spp.) and Myoida (20 and 4%, respectively; mainly Dreissena spp.). About one third of records, predominantly from Europe and the Americas, related to species that were non-native to the country of study. The majority of records originated from Asia (35%), with available evidence for 23 CICES classes, as well as Europe (29%) and North America (23%), where research was largely focused on 'biomonitoring'. Whilst the earliest record (from 1949) originated from North America, since 2000, annual output of records has increased rapidly in Asia and Europe. Future research should focus on filling gaps in knowledge in lesser-studied regions, including Africa and South America, and should look to provide a quantitative valuation of the socio-economic costs and benefits of ecosystem services shaped by freshwater bivalves.