In 2016, two earthquakes of magnitude greater than 6 and an earthquake of magnitude 6 occurred in Italy, in the Apennines. To document the slip distribution associated with the last earthquake of the sequence, the Norcia earthquake of Mw 6.6 (30/10/2016), we did a geodetic study combining InSAR, GPS, image correlation and field measurements. This earthquake comes from the breakdown of a normal fault system: the Monte Vettore fault system. Previous studies and field observations had revealed a displacement up to 2m on the main fault branch. The various geodetic data have made it possible to constrain movements in near and far fields of the Norcia earthquake. The image correlation confirmed that this rupture is segmented and comes from a relatively immature fault system. We have been able to characterize co-seismic displacements ranging from 30 to 200 cm on 8 branches of the fault, compatible at the first order with the field measurements. Despite these relatively numerous fault segments, the analysis of the length - displacement profiles indicates that the rupture can be divided into 2 major segments, which are likely to meet at depth. InSAR data show 2 lobes of deformation on the ascending track with values ranging from -40 cm to 20 cm, compatible with a main fault system and an antithetic fault, and 2 lobes on the descending track with displacements raging from -80 cm to 20 cm. Far-field GPS data indicate millimetric co-seismic displacements. Time series analysis suggests that a limited post-seismic relaxation occurred with a characteristic time of 10 days, which suggests that the mechanism at work is afterslip, maybe combined with poroelastic rebound. These co-seismic surface deformation data have been inverted to obtain a model of slip distribution on the buried fault. The geometry of the fault has been constrained from the focal mechanism and the fault trace at the surface: N153-40° SW. The calculated geodetic Mw is 6.66. The maximum slip obtained is of the order of 2.25m, with the presence of a single asperity located between 5 and 8 km depth. The model proposed here does not present any proof of segmentation in depth, despite the fact that the surface sliding is segmented.