It is usual to rank the participating countries in Olympic Games in accordance with the number of medals they have won. But it should be mentioned that achieving competition output (medals acquired) is possible only through utilization of input and appropriate processes. So, lower input should be considered along with higher outputs to evaluate more precisely and regularly. In this study, one of non-parametric techniques, namely data envelopment analysis (DEA) method, was used to design a model to evaluate the performance of participating countries in Olympic Games and to eliminate this problem. The model considered two inputs (GDP per capita and population) and three outputs (number of gold, silver and bronze medals acquired) to evaluate the performance of the participating countries. The data used to test the model related to the performance of countries in Athens Olympics (2004). The results of implementing this model showed that out of 73 countries participating in Athens Olympic Games which had achieved at least one medal, nine countries achieved maximum efficiency using this model.