This paper proposes a model predictive current control strategy based on optimal switching sequences for three-port converters connected to a permanent magnet synchronous machine which has application on hybrid systems. The optimal switching sequence is based on first-order mathematical techniques and an analysis of vector control for power converters. The strategy uses the principles of field oriented control to find reference currents that are taken by an outer model predictive control stage to control the machine. Then, the distribution of a redundancy vector is optimized on an inner MPC stage to control the third port of the converter. The proposed strategy was validated by simulation and compared with the classic current control based on finite control set. The results demonstrate high precision robust machine control with defined harmonic distortion decoupled from the third port operation, without sacrificing dynamic performance. In addition, the control method has low computational burden without harming the converter efficiency, making it feasible to implement in an experimental setup or real applications such as hybrid vehicles or storage systems.