Abstract: For a multivariable discrete-time stochastic system, a multiple models direct adaptive decoupling controller based on generalized minimum variance performance index is presented. It is composed of multiple fixed models and two adaptive models. The fixed models are used to cover the region where the system parameters jump, while the adaptive models are used to guarantee the stability and improve the transient response. For the fixed models, it utilizes the matrix pseudo-commutativity and quasi-Diophantine equation to design the generalized minimum variance controller. For the adaptive models, it adopts the stochastic system identification algorithm with optimal controller design method to identify the controller parameter directly. Then, through the choice of the weighting polynomial matrices, it not only realizes the dynamic decoupling control but also eliminates the steady state error and places the closed-loop poles arbitrarily. Finally, the global convergence is given. The simulation proves the effectives of the controller proposed.