This study aimed to forecast the prices of rice, milk, and palm oil in India using machine learning models. Historical price data from various regions, along with external factors like the ALPS Phase, were used to train and evaluate models such as Random Forest, XGBoost, and LSTM. After rigorous data preprocessing and feature engineering, the models were assessed using metrics like RMSE, R-squared, MAE, and MAPE. XGBoost emerged as the top performer overall, demonstrating superior accuracy in predicting commodity prices. However, due to the inherent volatility of rice prices across different states, the models faced challenges in accurately forecasting rice prices compared to milk and palm oil. The study generated 12-month price forecasts for each commodity, providing valuable insights into future price trends. The results demonstrate the effectiveness of machine learning models in capturing nonlinear patterns in agricultural price data. Accurate price forecasting can empower policymakers to implement targeted interventions, producers to optimize their production strategies, and investors to make informed decisions. Ultimately, the research tries to offer a reliable forecasting model that would help enhance the resilience of India's agricultural sector.