Analog circuits are of great importance in electronic system design. Analog circuit design consists of circuit topology design and component values design. These two aspects are both essential to computer aided analog circuit evolving. However, Traditional GA is not very efficient in evolving circuit component’s values. This paper proposed a hybrid algorithm HME-GA (GA with hyper-mutation and elitist strategies). The advantage of HME-GA is that, it not only concentrates on evolving circuit topology, but also pays attention to evolving circuit component’s values. Experimental results show that, the proposed algorithm performs much better than traditional GA. HME-GA is an efficient tool for analog circuit design. Evolutionary technology has been demonstrated to be very useful in computer aided analog circuit design. More potential of evolutionary methods on analog circuit design is waiting for exploring.