Skin cancer is one of the most commonly diagnosed cancers, with its incidence rapidly increasing and placing a significant burden on healthcare systems. Early detection is critical for improving patient outcomes, and recent advances in AI have shown promise in accurately identifying skin cancers. However, the high computational demands of AI models often limit their practical application, especially in resource-constrained environments. This study investigates the use of knowledge distillation to reduce the computational requirements of a skin cancer detection model while preserving diagnostic accuracy. Knowledge distillation involves training a smaller model to replicate the performance of a larger, more complex model, enabling it to function with fewer computational resources. In this study, a smaller skin cancer classification model was distilled from a more complex one, resulting in a model that was half the size of the original. The distilled model's area under the precision-recall curve (PR-AUC) nearly doubled, while the area under the receiver operating characteristic curve (ROC-AUC) saw a substantial increase, improving from 0.65 to 0.85 compared to the undistilled version of the same model. These findings suggest that knowledge distillation can successfully enable the generation of smaller, high-performing diagnostic models, making advanced skin cancer detection technologies more accessible in diverse healthcare settings.