This technical report describes the new Seasonal Prediction System developed at CMCC to perform seasonal forecasts operationally (CMCC- SPS3). A more realistic representation of the Climate System components such as the ocean, the sea ice, the snow cover, the soil moisture and the stratosphere is crucial to obtain reliable forecasts at the sub-seasonal to seasonal time-scale. This new Seasonal Prediction System currently operational at the Euro-Mediterranean Center on Climate Change was indeed developed with the aim of achieving enhanced predictive skill in a variety of different aspects. In comparison to the previous system (SPS2), the new model has a completely different dynamical core, based on the new CMCC Earth System Model (Fogli and Iovino, 2014). The new system features a better horizontal resolution of both the atmospheric and oceanic components, better representation of the stratosphere, more realistic initialization procedures for atmosphere, land, sea and ice modules and a larger ensemble size (50 members). Such improvements have a positive impact on the climate and on the predictive skill of the new system. After a brief description of each system component, the initialization strategy is discussed along with the main characteristics of the forecast system from a technical point of view. An analysis of its climate and of the forecasting skill is presented for the 24- year re–forecast period 1993–2016. The technical report is concluded with a preliminary attempt of comparison of the SPS3 overall performances with both SPS2 and other Seasonal Prediction Systems.