Masonry aggregates, which emerged as layouts of cities and villages became denser, make up historical centres all over the world. In these aggregates, neighbouring structures may share structural walls that are joined at the interfaces by mortar or interlocking stones. For instance, observations following the recent earthquakes in Italy and Croatia frequently revealed symptoms of separation and pounding at interfaces because of the out-of-phase behaviour of the units. The analysis of such building aggregates is complicated by the missing guidelines, as the scarce experimental data prevented the advances. Therefore, the objective of the project SERA AIMS (Adjacent Interacting Masonry Structures), included in the H2020 project SERA, was to create such data by testing an aggregate of two buildings under bidirectional dynamic excitation. The test unit was built at half-scale, with a two-storey building and a one-storey building. The buildings shared one common wall with the façade walls connected only by a layer of mortar, without interlocking stones. The floors were at different heights and had different beam orientations to facilitate the out-of-phase response. First, the shake-table test and main findings are briefly discussed. Second, blind-predictions coming from a dozen of research groups and consultancies are reported and compared with the experimental results. The comparison of blind-predictions and experimental results was based on quantitative comparison of reported displacements and base shear-values, and qualitative comparison of observed damage mechanisms. Submissions were grouped according to modelling approaches and key modelling assumptions to derive conclusions on their impact on the results. The work presented in this paper confirmed that even when the material parameters are known, and all the participants start from the same starting point – uncertainties related to modelling unreinforced masonry aggregates are such that predicted results vary greatly both in terms in reported displacement, interface opening, and damage mechanisms. Additionally, when compared to experimental results, correct predictions were rare – leading to the conclusion that there are many remaining uncertainties that have to be addressed when modelling this widespread building typology.