Portfolio optimization which means choosing the right stocks based on the highest return and lowest risk, is one of the most effective steps in making optimal investment decisions. Deciding which stock is in a better position compared to other stocks and deserves to be selected and placed in one's investment portfolio and how to allocate capital between these stocks, are complex issues. Theoretically, the issue of choosing a portfolio in the case of minimizing risk in the case of fixed returns can be solved by using mathematical formulas and through a quadratic equation; but in practice and in the real world, due to the large number of choices in capital markets, the mathematical approach used to solve this MODEL, requires extensive calculations and planning. Considering that the behavior of the stock market does not follow a linear pattern, the common linear methods cannot be used and useful in describing this behavior. In this research, portfolio optimization using the gray wolf algorithm and the MARKOWITZ MODEL based on CO-GARCH MODELing has been investigated. The statistical population of the current research included the information of 698 companies from the companies admitted to the Tehran Stock Exchange for the period of 2011 to 2020. First, the optimal investment MODEL is presented based on the gray wolf algorithm, and After extracting the optimal MODEL, the efficiency of the gray wolf algorithm is compared with the MARKOWITZ MODEL based on CO-GARCH MODELing.