Tomatos are one of the most important crops in global agriculture, responsible for many of the world’s vegetable production.Tomatoes are one of the most important crops in global agriculture, responsible for many of the world’s vegetable production. However, various diseases and pests often affect tomato plant health and yield. Early detection of these problems is essential to avoid crop loss and to optimize agricultural practices. It helps farmers take action on time for healthier tomato production. This paper uses a deep learning approach to predict leaf diseases by building a deep Convolutional Neural Network (CNN) model. Many different data sets are used to train the proposed model, including healthy tomato leaves and ones affected by diseases. The dataset that is used contains 1 class of healthy images consisting of 154 images and 10 classes of plant disease images consisting of 2869 images for the detection of diseases such as Bacterial spot, early blight, late blight, leaf-mold, powdery mildew, septoria leaf-spot, spidermitted two-spotted, target spot, Mosaic virus, yellow leaf-curl virus. The data is collected from the plant village. Multiple layers of convolution and Max pooling are used to build the model. The model accuracy is found to be 96.25%.