Patients with type 2 diabetes (T2DM) are associated with a high prevalence of chronic kidney disease (CKD). Diabetes with CKD is associated with poor long-term outcomes and high costs due to CKD progression and complications. To address the disease burden of CKD progression among patients with T2DM, risk scores that predict the CKD progression to identify individuals with T2DM at increased risk for developing progressive CKD are needed. This study used data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial and its follow-up trial (ACCORDION) to develop time-varying Cox prediction models for CKD and its progression. Model performance was assessed by discrimination and calibration. Risk scores were generated from the prediction model following the age-standardized integer-based Framingham Risk Scores development algorithm. Data were split into a training set and a validation set by a ratio of 7:3 for internal validation. Patient-level data of HARMONY outcome clinical trial and CRIC study were used for external validation. The predictive models and risk scores improved the prediction accuracy and the application, which would be suitable for clinical decision support, healthcare resource allocation, and communication improvement between physicians and patients/families.