The Gas-Assisted Gravity Drainage (GAGD) process has been suggested to improve oil recovery in both secondary and tertiary recovery through immiscible and miscible injection modes. In contrast of Continuous Gas Injection (CGI) and Water-Alternative Gas (WAG), the GAGD process takes advantage of the natural segregation of reservoir fluids to provide gravity-stable oil displacement and improve oil recovery. In the GAGD process, the gas is injected through vertical wells to formulate a gas cap to allow oil and water drain down to the horizontal producer (s). In this paper, a field-scale compositional reservoir simulations were conducted to study the feasibility of the GAGD process to enhance the recovery of oil in synthetic and real oil reservoirs. The GAGD process was implemented through the 5th SPE comparative solution project model (SPE5) and the heterogeneous upper sandstone oil reservoir in the South Rumaila oil field, located in Southern Iraq. Four different gas mixtures were injected: carbon dioxide, flue gas, nitrogen to methane, and associated gas (AG). In the SPE5 model, it was investigated that CO2-AGD process is much better than other gas mixtures with respect to achieving the highest oil recovery because of the influential role of CO2-rock-oil interaction to enhance the recovery of oil in this homogeneous system. The flue gas and nitrogen plus methane mixtures had similar efficiency by obtaining approximate oil recovery. However, the results of south Rumaila field GAGD evaluation indicated that the associated gas has a slight higher oil recovery than other gas mixtures including CO2 because of its compatibility with the existing reservoir and fluid properties. More specifically, the oil recovery by the end of the perdition period through the CO2, AG, flue gas, and N2+CH4 were 73.88%, 74.25%, 74.13%, and 73.91%, respectively; whereas, oil recovery at the beginning of the prediction period is 66.8%. In addition, there are many other reservoir factors have significant impact on the process efficiency and led to change the efficiency of each of the four gas mixtures, such as heterogeneity. It can be concluded that the immiscible flooding of using associated gas for re-injection in the South Rumaila field has the same effect of using the CO2 with respect to achieving a promising oil recovery. Consequently, associated gas can be efficiently utilized for a pilot EOR project implementation in the Rumaila field as a cheap solvent alternative to the carbon dioxide.