The relevance of this study is determined by the need to improve the efficiency of classification processes in the chemical industry, which utilizes powdered catalysts and adsorbents in the particle size range of 10-60 μm. Existing separation methods, including mechanical sieving, gravitational and inertial devices, cyclones, and rotary classifiers, are limited in terms of selectivity, reliability, or operational complexity. As an alternative, a multi-vortex classifier design with coaxially arranged tubes is proposed. A multi-vortex system is formed in the annular channel of the device, where solid particles are separated from the gas flow under the action of inertial forces and settle in the classifier’s collection bin. The aim of the study is to perform numerical modeling of particle fractionation in the classifier and to evaluate fractional efficiency and hydraulic resistance under varying geometric parameters. The simulation was conducted in the ANSYS Fluent software environment using a steady-state three-dimensional approach, the k-ω SST turbulence model, and the Discrete Phase Model. The study investigated the effects of vortex diameter and slit opening ratio, which were varied in the ranges of 17,5-29 mm and 0-1, respectively. The inlet gas velocity was set to 12 m/s, and particle sizes ranged from 1 to 200 μm. To evaluate fractional efficiency, a particle trap condition was applied to the walls of the collection bin. It was found that increasing the vortex diameter enhances separation efficiency. A decrease in the slit opening ratio leads to a weakened vortex structure, reduced tangential gas velocity, and increased hydraulic resistance. The best performance was observed at vortex diameters of 27,5-29 mm and slit opening ratios of 0,4-1. The fractionalseparation efficiencyfor particlessized 20-25 μm exceeds 95%.