The RINTC research project (RINTC Workgroup, 2018), financed by the Italian Department of Civil Protection, is aimed at evaluating the seismic risk of buildings conforming to the Italian building code. Within the framework of this project, the attention has been recently focused on existing buildings, too. In this study, case-study structures, representative of the existing residential reinforced concrete (RC) building stock in Italy, are analyzed. These structures are three-storey buildings with compact rectangular plan, and they have been defined through a simulated design process, in order to represent two types of buildings, namely designed for gravity loads only during 1970s (gravity load designed, GLD) or for moderate seismic loads during 1990s (seismic load designed, SLD). GLD buildings are assumed to be located in three different sites, namely Milan, Naples and Catania, in increasing order of seismic hazard. SLD buildings are assumed to be located in L'Aquila. The assumed design typologies are consistent with the seismic classification of the sites at the assumed ages of construction. The presence of typical nonstructural masonry infill walls (uniformly distributed in plan as external enclosure walls) is taken into account, assuming three configurations along height, namely "bare" (without infills), uniformly infilled and "pilotis" (without infills at the bottom storey) buildings. Two 554 COMPDYN 2019 7 th ECCOMAS Thematic Conference on Computational Methods in Structural Dynamics and Earthquake Engineering M. Papadrakakis, M. Fragiadakis (eds.) (not code-based) Limit States are investigated, namely Usability-Preventing Damage, corresponding to an interruption of the building use, and Collapse. RC elements are modelled with a lumped plasticity approach, through an empirical-based macromodel. The possible occurrence of shear failures in columns is taken into account through a preliminary classification of the expected failure mode (flexure-or shear-controlled, in the latter case prior to or following flexural yielding) and, if needed, a modification of the backbone of the nonlinear moment-chord rotation response, through empirical models providing the expected deformation capacity at shear and axial failure, the latter meant as the (initiation of) loss of axial-load-carrying-capacity. The nonlinear response of beam-column joints is modelled, too, with a "scissors model" based on concentrated springs representing the nonlinear response of the joint panel, at the intersection of beams' and columns' centerlines, through a preliminary evaluation of the expected failure mode (i.e. prior to or following yielding of adjacent beam/column elements). Materials properties are provided by literature studies, consistent with the age of construction of the buildings. The in-plane response of infills is modelled, taking into account the presence of openings, too. Modeling should be considered as simplified and, from some points of view, still preliminary, since advances are foreseen within the project in order to capture...