Based on the detailed analysis on peculiarity of DG, this paper establishes a multi-objective optimal model with DG by minimizing the electricity purchase costs, power loss and investment costs and maximizing voltage stability margin, which can reflect the DG planning layout practically and scientifically. On the basis, the super-efficiency data envelopment analysis is employed to determine the appropriate weights among the three objective functions, and in this way the multi-objective optimization problem is transformed into a single-objective programming one. In terms of the defects of traditional firefly algorithm, which is easily premature, has slow speed of convergence, and excessively relys on control parameters, this paper proposes a modified firefly method by introducing chaos search strategy and overall thought into firefly algorithm and applies it to solve the DG planning problem. Finally, the practicality and adaptibility of the proposed algorithm are illustrated by experiments.