Accurate forecast models of the near‒Earth radiation belt environment are of great importance to satellite operators and engineers, as the charged particles can be damaging to satellite hardware. The A Geosynchronous Radiation‒belt Electron Empirical Prediction (GREEP) model is presented in this study and utilized to predict the daily average of 1.8–3.5 MeV relativistic electron fluxes at Geostationary Earth Orbit (GEO), based upon propagated solar wind velocity (Vsw) and solar wind number density (n) measurements. The occurrence distributions of Vsw and n are used to normalize the distribution of flux with Vsw and n, respectively. A power law fit is applied to each probability distribution function, and the fitted values are dynamically combined with recent flux measurements from GEO and with fluxes from previous solar rotations to predict the future radiation belt environment at GEO. The model outperforms recent models in terms of forecast score for 1 day predictions, and it is found that the distribution of flux as a function of Vsw generally provides better predictions during the descending phase of the solar cycle than the distribution of flux as a function of density. The latter performs better near solar maximum. The GREEP model performs best in terms of forecast score also during the declining phase, and it provides further evidence of the importance of solar wind velocity in controlling high‒energy electron flux. During the descending phase, relativistic fluxes are typically the highest, and thus most damaging to spaceborne equipment, and pose a greater risk to humans in space.