Abstract In this paper, an innovative adaptive dynamic programming (ADP) algorithm with fast convergence speed is designed for the optimal containment control problem of discrete‐time linear multi‐agent systems. Precisely, a quadratic input energy cost function, including local containment error information and actuator information in the neighborhood, is designed for each follower. Solving the stationary condition of the cost function, the optimal containment controllers are obtained. Traditional ADP methods use actor–critic neural networks to approximate optimal costs and control strategies, it is time‐consuming to solve large‐scale multi‐agent problems due to the computational complexity of neural networks. In order to seek faster convergence speed of optimal containment control without knowing the model information, the fast ADP algorithm framework is designed, it is proved theoretically that the convergence speed is determined by some configurable parameters, and the whale optimization algorithm is employed to globally optimize the parameters of given spaces to derive the optimal configuration. Finally, numerical simulation results are given to verify the effectiveness of the designed algorithm.