This study addresses the pressing need for optimized solar power systems in the context of climate change concerns. Focusing on Maximum Power Point Tracking (MPPT) techniques, the research evaluates various models to enhance energy generation in solar systems under fluctuating solar irradiation conditions. The Adaptive Neural-Fuzzy Inference System (ANFIS) is chosen for its responsiveness, but designing an efficient ANFIS-MPPT system requires precise training data. The study introduces a novel approach, combining ANFIS with Gene Expression Programming (GEP), aimed at optimizing the reference maximum power output using solar irradiance and temperature as input parameters. The integration was tested on a boost converter via Matlab/Simulink simulations, which reveals the GEP-ANFIS double diode model's exceptional 99.84% efficiency under high solar irradiation. This underscores the substantial potential of GEP-ANFIS for improving solar power efficiency and MPPT performance in diverse environments, contributing to the advancement of solar energy utilization.