Soil results from the continuous evolution of parent material under natural constraints: climate, biology and relief (Jenny, 1961). This pedogenesis is responsible for soil spatial distribution. Thus a better understanding of soil distribution requires an integration of past pedological processes through modeling. In addition, human activities and climate change are acting as pedogenetic factors on recent soil evolution; their consequences are however poorly known and quantified. To be able to predict soil evolution in the future depending on different climate scenario or land use management, modeling approaches are necessary. We proposed here a review of existing models enable to predict pedogenesis. The literature proposed different ways of modeling: geostatistic (McBratney et al., 2003), deterministic (e.g. Phillips et al. 1996) or mechanistic (e.g. Jacques et al., 2002). In this review, we are discussing the effect of different time scale in pedogenesis to choose the more adapted model. We propose thus to distinguish a long (103-104 yr) and a short (101-102 yr e.g. human impact) time scale when considering pedogenesis. For the long time scale, model focus on the understanding of major processes of soil formation resulting in the actual soil three-dimensional spatial distribution through mass balance approaches(Brimhall et al., 1985; Minasny and McBratney, 1999). In this case, only main soil processes are considered. Brimhall's approach (Brimhall et al., 1985; 1988; 1991) requires a good characterization of the parent material and hardly take into account erosion process. This approach was mainly used for tropical weathering profiles like for example laterisation. Minasny approach (Minasny and McBratney, 1999; 2001) mainly considers erosion effect at the landscape scale. For short time scale, existing models mainly used for environmental concern could be used to predict recent pedogenesis. These approaches consist in water flow model coupled with particle and solute transport (Valles et al., 1990a and b; Jacques et al., 2002). The use of a geochemical model, like PhreeQc (Parkhurst and Appelo, 1999), allows considering the kinetic approach of the soil processes indeed in most temperate soils equilibrium is rarely meet. These different approaches were however rarely compared on tangible cases. References: Brimhall, G.H., et al. 1985. Economic Geology 80: 1227-1254. Brimhall, G.H., et al. 1988. Nature 333: 819-824. Brimhall, G.H., et al. 1991. Geoderma 51: 51-91. Jacques, D. et al. 2002. Computational methods in water ressources, 1: 555-562. Jenny, H. 1961. Soil Science Soc. Am. Proceedings 25: 385. McBratney, A.B., et al. 2003. Geoderma 117:3-52. Minasny, B., and A.B. McBratney. 1999. Geoderma 90: 3-21. Minasny, B., and A.B. McBratney. 2001. Geoderma 103: 161. Parkhurst, D.L., et Appelo 1999 Water Ressources Investigation Report 99-4259, Denver, Co USA 312 p Valles, V., et al. 1990a. Science du sol 28: 137-147. Valles, V., et al. 1990b. Science du sol 28: 149-159.