Extracting an optimal set of parameter values for a MOS device is great importance in contemporary technology is a complex problem. Traditional methods of parameter extraction can produce far from optimal solutions because of the presence of local optimum in the solution space. Genetic algorithms are well suited for finding near optimal solutions in irregular parameter spaces. In this study*, We have applied a genetic algorithm to the problem of device model parameter extraction and are able to produce models of superior accuracy in much less time and with less reliance on human expertise. MOS transistor's parameters have been extracted and optimized with genetic algorithm. 0.35μm fabricated by C35 process have been used for the results of experimental studies of parameter extraction. Extracted parameters' characteristic data results have been compared with measurement results. Different values of parameters of genetic algorithm, such as population size, crossover rate , and generation size are compared by different tests.