The purpose of this work is to contribute to a better control of the crystallization process which occurs in a supercritical medium, especially during the Supercritical AntiSolvent (SAS) process. It also aims to improve the prediction of crystal habit, thanks to the use of the molecular modeling software GenMol. The first part of the work was devoted to the crystal modeling of the two main forms of sulfathiazole in vacuo, considering Hartman’s attachment energy formalism. The second part considers solvent–crystal interactions throughout adsorption simulations to investigate the effect of growth environments on crystal habits. Lastly, modeling predictions were compared with grown crystals of sulfathiazole, observed after recrystallization with the SAS process from acetonitrile, acetone, tetrahydrofuran and acetic acid solutions. These comparisons demonstrated good predictions of crystal habit taking into consideration the growth environment. Neither carbon dioxide (antisolvent of the SAS process) nor acetonitrile leads to a modification of the isometric, in vacuo predicted habit of both forms. Acetone and tetrahydrofuran adsorb preferentially on some identified faces and lead to flat, leaflike, or tabular crystals. Acetic acid adsorbs on every one of the faces and hinders the phase transition to a more stable form, thus leading to crystals of the least stable, kinetically favored form I. Experimental observations were also rationalized by considering the different possible crystallization pathways, in particular Crystallization by Particle Attachment and Droplet Drying mechanisms occurring in the SAS process. This work confirms that solvent nature is one of the key elements to consider in order to better control the characteristics of particles grown using the SAS process and provides a new method to help to control it.