The SERA project AIMS (Seismic Testing of Adjacent Interacting Masonry Structures) comprised a blind prediction competition related to a shake-table test of an aggregate of two buildings under two horizontal components of dynamic excitation. Each group was provided with a complete set of construction drawings, material properties, testing sequence and the list of measurements to be reported. This paper deals with the prediction submitted by the EPFL team. The prediction was based on an equivalent frame model with a newly developed macroelement able to simulate both the in-plane and out-of-plane behaviour of unreinforced masonry piers. This paper describes the model and presents results of the pre-diction and preliminary results of the post-diction. For the post-diction the same model as for the pre-diction was used and only the seismic input was updated to match the actually applied shake table motion. The paper concludes with future work needed to calibrate the model for the post-diction.