Abstract We report on the results of about 9 months of gravimetric recordings acquired at Mt. Somma-Vesuvius (SV) volcano (Southern Italy) with the new generation relative gravimeter gPhoneX#116 (gPh#116), which is a gravimeter specifically designed for continuous gravity recording. We also present the outcomes of an intercomparison experiment of the gPhone#116 conducted at the J9 gravity observatory in Strasbourg (France). In this intercomparison, we were able to check the scale factor of the meter with a high degree of precision by means of an intercomparison with 2 superconducting gravimeters (SGs) and a FG5-type absolute ballistic gravimeter. Multiple calibration approaches allowed us to validate the manufacturer's original calibration constants to a level of 1% accuracy and 0.1% precision. Moreover, we carried out a comparative study of the noise level of the gPh#116 with respect to the SGs and other spring meters routinely used in both prospecting and time-lapse gravimetry. It turns out that gPh#116 exhibits lower levels at hourly time-scales than other compared spring gravimeters (Graviton, gPhone#054, Scintrex-CG5). It was also possible to carry out a detailed study of the instrumental drift, a crucial topic for reliable monitoring of the long-term gravity variations in active volcanic areas. In fact, a challenge in time-lapse gravimetry is the proper separation of the instrumental variations from real gravity changes eventually attributable to recharge or drainage processes of magma or fluids in the feeding systems of active volcanoes. A negative finding coming out from the intercomparison is that, even when applying the tilt correction, the gravimetric residuals obtained with the gPh#116 are an order of magnitude larger and quite inconsistent with those obtained with co-located superconducting gravimeters. We guess this problem could be overcome by installing the gravimeter on an auto-levelling platform. From the analysis of the gravity records, a reliable tidal gravity model was derived, which we believe will help to improve the accuracy of volcano monitoring, as it will allow appropriate correction of tidal effects for both relative and absolute gravity measurements acquired in the area. Two further interesting elements arose from our study: (1) a peculiar cavity effect of the SV underground laboratory that seems to influence the tilt change; (2) the small residual gravity signals are time correlated with the rainfall peaks and are compatible with gravity decreases induced by increases in soil moisture above the gravimeter.