Abstract: Secondary-ion batteries, such as lithium-ion (LIBs) and sodium-ion batteries (SIBs), have become a hot research topic owing to their high safety and long cycling life. The electrode materials for LIB/SIBs need to be further developed to achieve high energy and power densities. Anode/cathode active materials based on their alloying/dealloying with lithium, such as the anode materials of silicon, phosphorus, germanium and tin, and the cathode material of sulfur, have a high specific capacity. However, their large volume changes during charging/discharging, the insulating nature of phosphorus and sulfur, as well as the shuttling of polysulfides in a battery with a sulfur cathode decrease their specific capacity and cycling performance. The formation of dendrites in anodes during the deposition/dissolution of Li and Na leads to severe safety issue and hinders their practical use. Carbon materials produced from abundant natural resources have a variety of structures and excellent conductivity making them suitable host frameworks for loading high specific capacity anode/cathode materials. Recent progress in this area is reviewed with a focus on the factors affecting their electrochemical performance as the hosts of active materials. It is found that the mass loading of the active materials and the energy density of the batteries can be enhanced by increasing the specific surface area and pore volume of the carbon frameworks. Large volume changes can be efficiently accommodated using high pore volume carbon frameworks and a moderate loading of the active material. Suppression of the shuttling of polysulfides and therefore a long cycling life can be achieved by increasing the number of binding sites and their binding affinity with polysulfides by surface modification of the carbon frameworks. Dendrite growth can be inhibited by a combination of a high specific surface area and appropriate interface modification. Rate performance can be improved by designing the pore structure to shorten Li⁺/Na⁺ diffusion paths and increasing the electrical conductivity of the carbon frameworks. DFT calculations and simulations can be used to design the structures of carbon frameworks and predict their electrochemical performance.