Soluble gas stabilization (SGS)1 technology is a pre-step process of dissolving carbon dioxide (CO2) into the muscle food before packaging. This treatment relies on the bacteriostatic effect of CO2. However, the process is time-consuming and thus inefficient for industrial line processing. An accurate coupled numerical model can pave the way toward more efficient methods, suitable for large-scale industrialization. A computational fluid dynamics (CFD) model in conjugation with the gas diffusion is used to simulate the SGS process, estimating the CO2 dissolution in the food product, specifically inside salmon fillet as a function of time. The CO2 diffusion in salmon inside a chamber with an initial gas concentration in the headspace and constant temperature is simulated using COMSOL Multiphysics. To calculate the effective diffusion coefficient of CO2 in salmon, two phenomena of pure diffusion of CO2 and the density-driven natural convection induced by CO2 dissolution in the food are simulated. Food is modeled as a porous media, where CO2 is both transferred and dissolved. The porosity is calculated based on the histological sections of the salmon fillet with the assumption that CO2 dissolves in the liquid phase of water and fat inside the pores. Furthermore, the simulations are validated by experimental results and simplified analytical solutions. The simulation of CO2 diffusion in salmon showed good consistency with the experimental results in terms of the dissolved gas in food over time. This method not only makes it possible to simulate the SGS process for further investigations but prevents expensive time- and cost-consuming experiments to calculate the effective diffusion coefficient of CO2 in food at different experimental conditions. An accurate validated numerical model provides the opportunity to examine different design configurations and possibilities, aiming for an optimized design for a full-scale SGS technology.