In this paper, a novel model predictive control (MPC) scheme is proposed for field-oriented control of permanent magnet synchronous motor fed by voltage-source inverter, in which a hybrid control set (HCS) including active-vectors with duty cycle and virtual vectors with duty cycle can be established. Since more vectors with different angles and duty cycles are available, the proposed approach is capable of adjusting the magnitude and angle of the output vector finely. However, the duty cycle of vectors in HCS is countless, so the total of elements in HCS is actually infinite. In order to select the optimal vector from HCS, we transform the original MPC optimization problem. By means of finite enumerations and evaluations online, the optimal solution can be obtained, which includes both angle and duty cycle of the output vector. In addition, a simplified solving algorithm is also presented, which can further reduce the computational cost. Finally, the feasibility and effectiveness of the proposed approach can be validated by experimental results.