Background: The rapid global spread of SARS-COV-2 forced governments to implement drastic interventions. The existence of a large but undetermined number of mild or non-symptomatic but infectious cases seems to be involved in the rapid spread, creating a high level of uncertainty due to the difficulty to measure them, and difficulty for epidemiologic modelling. Methods: We developed a compartmental model with deterministic equations, that accounts for clinical status, mobility, r heterogenous susceptibility and non-pharmaceutical interventions . The model was calibrated using data from different regions and we used it to predict the dynamic in Buenos Aires Metropolitan Area (AMBA). Results: The model adjusted well to different geographical regions. In AMBA the model predicted 21400 deaths at 300 days, with 27% of the population in the region immunized after the first wave, partly due to the high incidence of asymptomatic cases. The mobility restriction is approximately linear, with any restriction bringing a positive effect. The other interventions have a combined effect of 27% reduction in infection rates. Conclusion: Our research underlines the role of asymptomatic cases in the epidemics' dynamic and introduces the concept of susceptibility heterogeneity as a potential explanation for otherwise unexplained outbreak dynamics. The model also shows the big role of non-pharmaceutical interventions both in slowing down the epidemic dynamics and in reducing the eventual number of deaths. The model results are closely compatible with observed data.