Diabetes causes a disease in the eye known as diabetic retinopathy. It is when high levels of glucose damage the blood vessels of the retina. These blood vessels may swell or be closed and thus prevent blood from flowing through them. Sometimes, an excess number of new vessels develop on the retina. All these changes can take away your vision. There are two main stages of diabetic eye disease: NPDR and PDR. NPDR is early, featuring leaking blood vessels, swelling, and blurry eyesight among others. PDR is advanced: which means that new blood vessels growing cause bleeding, scarring as well as possible retinal detachment with severe consequences for vision loss. India has about 77 million people living with diabetes and this number expectedly is supposed to rise to 125 million by 2045 [1].This research explores the application of deep learning architectures in the binary classification of Diabetic Retinopathy. Utilizing the 2015 Colored Resized dataset and the Diagnosis of Diabetic Retinopathy dataset, both available on Kaggle. MobileNetV2 achieved a highly commendable accuracy of 95.8%, while Xception attained an accuracy rate of 89.4%.