Parameter mismatch will result in inaccurate motion behavior prediction, which will reduce the performance of the predictive algorithm. To increase the system’s robustness against dynamic changes of parameters and disturbances, a disturbance preview based multistep finite control set model predictive current control (DP-MPCC) method is created for a permanent magnet synchronous motor (PMSM) system controlled by a two-level voltage-source inverter (2L-VSI), in which the disturbance observer is implemented by the reduced-order generalized proportional integral observer (Reduced-order GPIO). This strategy can prepare heavy computations for the long horizons optimization problem offline without increasing the computational burden. Additionally, the multistep prediction algorithm can now be real-time embedded with the estimation of the disturbance sequence and the dynamic parameter changes in the PMSM, which greatly enhances prediction accuracy and control effectiveness. Simulated results are shown, which support the effectiveness of the proposed methodology.