In practice, Computational Fluid Dynamics (CFD) based on the Navier-Stokes-Fourier equation is applied in the continuum regime, while Direct Simulation Monte Carlo (DSMC) based on the Boltzmann equation is applied in the rarefied regime. However, many aerodynamic engineering problems, such as re-entry vehicles, involve multi-scale rarefied gas flows where the density significantly varies locally. For multi-scale rarefied gas flows, CFD has limitations in terms of accuracy, while DSMC has limitations in terms of efficiency. Recently, alternative particle Monte Carlo methods have been proposed to bridge the gap between CFD and DSMC. These include the Bhatnagar-Gross-Krook (BGK) method and the Fokker-Planck (FP) method. However, there is a scarcity of comprehensive comparative studies to assess their performance. The present paper provides consistent comparison results among CFD, DSMC, BGK, and FP for rarefied gas flows.