In this work, a stochastic reactor model (SRM) is presented that bridges the gap between multi-dimensional computational fluid dynamics (CFD) models and zero-dimensional models for simulating spark-ignited internal combustion engines. The quasi-dimensional approach calculates spatial temperature and composition of stochastic “particles” in the combustion chamber without defining their spatial position, thus allowing for mixture stratification while keeping computational costs low. The SRM simulates flame propagation using a three-zone combustion model consisting of burned gas, flame front, and unburned gas. This “flame brush” approach assumes a hemispherical flame front that propagates through the cylinder based on estimated turbulent flame speed. Cycle-averaged turbulence intensity (u’) is used in the model, calibrated using experimental data. Through the use of a kinetic mechanism, the model predicts key emissions such as CO, CO2, NO, NO2, and HC from both port fuel injection (PFI) and gasoline direct injection (GDI) engines, the latter through the implementation of a simplified spray model. Experimental data from three engines, two GDI and one PFI, were used to validate the model and calibrate cycle-averaged u’. Across all engines, the model was able to produce pressure curves that matched the experimental data. In terms of emissions, the simplified chemical kinetics mechanism matched trends of the experimental data, with the PFI results having higher accuracy. Pressure, burned fraction, and engine-out emissions predictions show that the SRM can reliably match experimental results in certain operating ranges, thus providing a viable alternative to complex CFD and single zone models.