Optimization methods are applied to discover a near optimal or optimal solution for any distinguished problem. Many researchers have applied different optimization techniques on microgrids for cost optimization. In this paper, an improved multi-verse optimizer algorithm is proposed for generation cost minimization in microgrids. Two modifications are done in the original algorithm for solving local optima problem and improving exploration and exploitation process. Local optima problem is solved using average positioning and universe position updating equation is improved by hybridizing it with sine�cosine algorithm. Thus, the simulation results reported by the investigated algorithms show that the proposed algorithm outperforms the other algorithms in minimizing generation cost and reducing computation time. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.