The prediction of stock market trends is still difficult due to its complex volatility and its vast range of factors affecting price variations. Therefore, the traditional models keep on failing; to detect the structures in financial data more appropriate approaches are required. This paper introduces a hybrid model using genetic algorithms (GA), combining multiple machine learning techniques to provide accuracy in the prediction of stock prices. By the combination of support vector machines, decision trees, and neural networks, a hybrid model accumulates all the strengths of algorithms to adapt to different stock data patterns. This hybrid model leverages GA for feature selection, optimizes parameters, and fine-tunes ensemble weights to find an effective fitting model that handles all the fluctuations in the markets. Tests reveal that this hybrid model boosted by a genetic algorithm (GA) is more accurate than traditional models, making it a valuable tool for stock forecasting.