In the present paper, we consider multiobjective two-level nonlinear integer programming problems (MOTLNLIPPs) in which the decision maker (DM) at each level controls his own integer decision variables to optimize multiple objective functions. Various approaches for multi-level programming problems could exist according to situations which the DMs are placed in. In this paper, it is assumed that the DMs have motivation to cooperate with each other and they have own fuzzy goals with respect to their multiple objective functions and partial information on their preferences among them. Under the situation, we propose an interactive fuzzy programming technique through genetic algorithms for MOTLNLIPPs to obtain a satisfactory solution for the DMs. Furthermore, the feasibility of the proposed method is shown by applying it to an illustrative numerical example.