The optimal operation of active distribution systems in the presence of private renewable-based entities is one of the primary challenges of future power networks. In this regard, developing a practical framework to deal with this kind of issue is essential. Hence, in this paper, a novel bi-level stochastic programming approach is presented for optimal energy and reserve scheduling of the active distribution system in the presence of different eco-efficient autonomous players. In the proposed model, the distribution system operator, as a leader, attempts to minimize its total operating costs. At the same time, the renewable-based microgrid owner, as an independent follower, tends to maximize its profit from exchanging energy and reserve with the distribution system operator. The suggested scheme is a non-linear bi-level problem which is transformed into a non-linear single-level problem through Karush–Kuhn–Tucker conditions. In order to find the global optima, the non-linear single-level problem is linearized by utilizing the Big-M method. Finally, to investigate the effectiveness of the provided model, it is tested on the modified IEEE 15-Bus active distribution system under different cases and scenarios. Obtained results indicate that the operation cost of the distribution system operator can be reduced up to 134.09$, from 10710.11$ to 10576.02$, and the profit of the microgrid owner can be increased significantly 906.93$, from 659.455$ to 1566.39$, by considering both environmentally friendly units, IPL and CAES.