This paper presents a model predictive control strategy for an optimal asymmetric multilevel inverter with floating H-bridge cell. Nowadays, multilevel converters are a very attractive alternative for the efficient handling of power electrical systems, compared to the traditional two-level inverters. The topology is composed by two stages, a main cell composed by a three-phase two level inverter and a secondary cell, composed by one H-bridge per phase. H-bridge cell has a floating capacitor instead of an isolated DC source. A Finite Control Set Model-based Predictive Control (FCS-MPC) strategy is proposed. The scheme is modified to achieve a better performance of the complete system and reducing the calculation. Main cell is controlled by Space Vector Control (SVC) and implemented by a simple mathematical technique. This result and the description of a cost function, are used to control the load current and the voltages of the floating capacitors for an optimal asymmetry. The performance of the proposed control strategy is evaluated through simulations and experiments.10