India is a country where major source of living and economy is through agriculture and agricultural industry. The cultivated area of Karnataka is 9.85 million comprising 51.4% of the total geographical area. Prediction of prices in agricultural commodity has always been a major problem for the farmers. Commodities like vegetables and pulses prices in the market fluctuate wildly these days. Commodity market prices can be predicted using machine learning techniques. This study forms the basis for a number of research papers. Although there doesn't seem to have reliable technique for predicting the market prices of commodities. Agriculture commodity prices in India fluctuate significantly due to differences in climatic conditions, regional supply, and market demand, making it challenging for farmers and traders to plan effectively. The proposed model uses region-specific data, such as commodity type, market location, and current price, to forecast prices for commodities from January to December. In order to estimate the prices of Agri-Horticultural commodities, including fruits, vegetables and pulses, in the Karnataka regions, this study proposes an AI-driven price prediction system that employs machine learning techniques. To improve model performance in prediction, data preprocessing procedures like cleaning, normalization, and feature encoding were carried out. Multiple regression and ensemble learning algorithms were implemented and compared, with Random Forest showing the best accuracy. The dashboard enables users to interactively choose commodities, regions, and forecast periods, offering both actual market data comparison and predictive analytics. The model is operated via an intuitive Streamlit interface. Performance evaluation metrics like as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE) and R2 can be used to assess the model’s correctness.