We perform a robust estimation of the Phillips curve in the euro area using a battery of 630 theory-driven models. We extend the existing literature by adding model specifications, taking into account the uncertainty in the measurement of variables and testing for potential non-linearities and structural changes. Using Dynamic Model Averaging, we identify the most important determinants of inflation over the sample. We then forecast core inflation 12 quarters ahead and present its probability distribution. We compare the distribution of forecasts performed in recent years, and we assess, in a probabilistic manner, the convergence towards a sustainable path of inflation.