This paper proposes two hybrid versions of the generalized normal distribution optimization (GNDO) and sine cosine algorithm (SCA) for global optimization. The proposed hybrid methods combine the excellent characteristics of the GNDO and SCA algorithms to enhance the exploration and exploitation behaviors. Moreover, an additional weight parameter is introduced to further improve the search ability of the hybrid methods. The proposed methods are tested with 23 mathematical optimization problems. Our results reveal that the proposed hybrid method was very competitive compared to the other metaheuristic algorithms.