Neutral point potential balance is an inevitable challenge in multilevel converters, as an imbalance can affect the lifespan of the DC-link capacitors. To achieve neutral point potential balance and enhance system reliability, this paper proposes a novel finite control set model predictive control scheme for switched reluctance motor (SRM) based on T-type three-phase four-leg converter. By using Kirchhoff’s current law, the neutral point potential at the next time step is predicted. Furthermore, based on the Le-Huy model, the torque at the next moment is estimated. The cost function includes torque ripple and the DC-link capacitor potential difference terms. By adjusting the appropriate weight coefficients, the neutral point potential balance can be achieved while reducing torque ripple. Meanwhile, to reduce computational burden, the number of candidate voltage vectors and corresponding switching states is optimized. The feasibility of the proposed scheme is validated on a 750W 12/8 SRM.