Capacitive deionization is an emerging and rapidly developing electrochemical technique for water desalination across the globe with exponential growth in publications. There are various architectures and materials being explored to obtain utmost electrosorption performance. The symmetric architectures consist of the same material on both electrodes, while asymmetric architectures have electrodes loaded with different materials. Asymmetric architectures possess higher electrosorption performance as compared with that of symmetric architectures owing to the inclusion of either faradaic materials, redox-active electrolytes, or ion specific pre-intercalation material. With the materials perspective, faradaic materials have higher electrosorption performance than carbon-based materials owing to the occurrence of faradaic reactions for electrosorption. Moreover, the architecture and material may be tailored in order to obtain desired selectivity of the target component and heavy metal present in feed water. In this review, we describe recent developments in architectures and materials for capacitive deionization and summarize the characteristics and salt removal performances. Further, we discuss recently reported architectures and materials for the removal of heavy metals and radioactive materials. The factors that affect the electrosorption performance including the synthesis procedure for electrode materials, incorporation of additives, operational modes, and organic foulants are further illustrated. This review concludes with several perspectives to provide directions for further development in the subject of capacitive deionization. PRACTITIONER POINTS: Capacitive deionization (CDI) is a rapidly developing electrochemical water desalination technique with exponential growth in publications. Faradaic materials have higher salt removal capacity (SAC) because of reversible redox reactions or ion-intercalation processes. Combination of CDI with other techniques exhibits improved selectivity and removal of heavy metals. Operational parameters and materials properties affect SAC. In future, comprehensive experimentation is needed to have better understanding of the performance of CDI architectures and materials.