Traditional computable general equilibrium (CGE) models based on the Armington assumption fail to capture the extensive margin of trade, and thereby underestimate the trade and welfare effects of trade opening. To address this problem, this paper introduces the Melitz (2003) theoretical framework with firm heterogeneity and fixed exporting costs into a global CGE model. Some illustrative simulations show that the introduction of firm heterogeneity improves the ability of the CGE model to capture the trade expansion and welfare effects of trade liberalization. Under the case of a global manufacturing tariff cut, the estimated gains in welfare and exports are more than double those obtained from a standard Armington CGE model. Sensitivity analysis indicates that model results are sensitive to the shape parameters of firm productivity distribution, suggesting the need for further empirical work to estimate the degree of firm heterogeneity.