This study investigates the application of Factor-Augmented Vector Autoregression (FAVAR) and Bayesian Vector Autoregression (BVAR) models for inflation forecasting. FAVAR models deal with high-dimensional data by extracting latent factors from extensive macroeconomic indicators, while BVAR models incorporate prior distributions to enhance forecast stability and precision in data-limited environments. Employing a comprehensive dataset of Uzbekistan-specific inflation determinants, we conduct an empirical assessment of both models, examining their predictive accuracy. Findings from this research aim to optimize inflation forecasting methodologies, providing the Central Bank of Uzbekistan with robust, data-driven insights for improved policy formulation.