Abstract: Owing to adopting the Matlab language, the genetic algorithms (GAs) are achieved and further, GAs Toolbox was set up. The developed GAs Toolbox supports both the binary and the floating-point encoding methods so that the performance of GAs is enhanced. The toolbox was used to test a series of nonlinear, multi-peak-value functions and demonstrates excellent performances. It is also found that the floating-point GA is superior to the binary GA and the traditional optimization methods in solution-finding rate and exactness.