This study introduces a novel stacking-based hybrid deep learning model for the detection of pneumonia, a critical task in the medical field that demands high accuracy and efficiency. The proposed model amalgamates the strengths of multiple deep learning models, including VGG16 and VGG19, to create an optimal prediction model. The stacking technique employed allows for the integration of the predictions of the different models, thereby enhancing the final prediction accuracy. The models were evaluated on a comprehensive dataset of chest X-ray images. The experimental results demonstrate that the proposed model achieves a remarkable accuracy of 96.875% in pneumonia detection. This research contributes significantly to the existing body of knowledge by presenting a unique approach to pneumonia detection and providing insights into the effectiveness of the proposed model. The potential for further improvements and integration into existing healthcare systems is also discussed.