This book chapter presents Model Predictive Control (MPC) strategies for Master-Slave parallel inverters in microgrids. The Master is a grid-forming inverter with an LC output filter, while the Slave is a grid-following inverter with an output LCL filter. The microgrid is also composed of linear and non-linear loads, as well as grid and line impedances. First, a Finite Control Set MPC (FCS-MPC) is described and validated by means of Hardware-in-the-Loop (HIL) results covering different operational conditions of the microgrid, including grid connection, islanded mode, and load variations. Even though FCS-MPC has a very good dynamic and transient performance, it inherently results in variable switching frequency for the inverters, increasing the harmonic content of the output voltages and currents and making the design of the output filter more complex. In order to overcome these issues, a Fixed Switching Frequency Model Predictive Control (FSF-MPC) is then presented for the same microgrid, being again validated through HIL for the same operational conditions as for the FCS-MPC. Finally, the performance of both MPC controllers is compared. The results presented in this chapter demonstrate the good performances of the proposed FCS-MPC and FSF-MPC approaches, such as fast dynamic response, multi-variable control, robustness to parametric uncertainties, and fixed switching frequency for the FSF-MPC.