In the process of large-scale analog integrated circuit design using miniaturization technology, there are some problems of conflicting performance indicators. An elitist guided multi-objective fireworks algorithm with difference variation(GMOFWA-DV) is proposed in this paper. Particle evolution information is used to guide the explosion and improve the efficiency of the algorithm. In order to enhance the information exchange between particles and facilitate the application of the guidance strategy, the mutation and cross operators in the differential algorithm are adopted in variation of the GMOFWA-DV. Simulation experiments of the comparision of the designed algorithm with 3 others are conducted. The experimental results verify the effectiveness of the proposed algorithm. The application of the GMOFWA-DV in the practical engineering application of CMOS analog integrated circuit design parameter optimization can reduce the development cycle of analog integrated circuit design.