The robust model predictive control (MPC) of systems with stable and integrating modes is addressed. The approach proposed here extends the method presented in Odloak in 2004 that can only be applied to open-loop stable systems. Here, the robust controller is developed assuming that there is model uncertainty in both stable and integrating parts of the system. The method considers a modified cost function that turns the infinite output horizon MPC globally convergent for any finite input horizon. The controller is based on a modified version of the state-space model utilized by Carrapico and Odloak in 2005 to develop a nominally stable MPC for systems with stable and integrating modes. The approach considers the inclusion of feasible cost contracting constraints in the control optimization problem, taking into account the annulment of the integrating modes to assure a bounded infinite horizon cost. A simulation example is included to illustrate the performance and robustness of the proposed approach and to demonstrate that the controller can be implemented in real applications. (c) 2007 American Institute of Chemical Engineers.