Pneumonia is one of the acute respiratory illnesses that affects the lungs, namely the alveoli, it is a dangerous respiratory infection that can result in major health issues. Deep learning models have recently demonstrated encouraging outcomes in medical image analysis applications, such as the diagnosis of pneumonia. The usefulness of two well-known deep learning models, You Only Look Once (YOLOv5) and Convolutional Neural Network (CNN), in identifying pneumonia from chest X-rays are investigated in this study. Experimental results on a chest X-ray image data set show that the YOLOv5 and CNN models can effectively detect cases of pneumonia. The YOLOv5 model achieved a significantly higher precision of 0.97, surpassing the CNN model's precision of 0.91. The YOLOv5 model demonstrated superior performance in accurately classifying positive cases and a high percentage of predicted positive results were validated. Further research and validation in larger and more diverse datasets is recommended to strengthen the generalizability and reliability of these models in clinical practice.