The most common deadliest disease is skin cancer if not treated and identified at early stages. Current estimates state one in five people in America are likely to have skin cancer over their entire lifetime. With the help of CNN models ResNet and EfficientNet, this study classifies the first skin cancer among the stages as Benign and Malignant. For further clarification, it also classifies the skin lesion among the 7 different classes among actinic keratoses (akiec), basal cell carcinoma (bcc), benign lesions of the keratosis type (bkl), dermatofibroma (df), melanoma (mel), melanocytic nevi (nv), and vascular lesions (vasc) with help of VGG16, AlexNet, DenseNet, and MobileNet models. All the mentioned models were exclusively trained and tested on HAM10000 dataset. Overall, the maximum accuracy for binary classification is 86% and for multiclass it is 84%.