A sliding mode controller is designed to compensate for the flexibility uncertainties of a cable robot and improve its tracking performance. Of the most significant sources of these uncertainties are the elasticity of the cables and the flexibility of the joints. A favorable approach to improve the accuracy of the system is first to model the cable and joint flexibilities and then convert the model uncertainties into parametric uncertainties. Parametric uncertainties are the product of imprecise flexibility coefficients and are finally neutralized by a sliding mode controller. The flexibility in cables is modeled by considering the longitudinal vibration of the time-varying length cables. A simulation study is carried out to confirm the presented model and the positive effect of the designed controller. Then the impact of these uncertainties on the dynamic load carrying capacity (DLCC) of the robot is examined and compared for different cases. Finally, experimental tests are conducted on the IUST (Iran University of Science and Technology) cable-suspended robot to validate the presented theories and simulation results.