Background The present stage of development of potato genetics and breeding is characterized by an intensified activity of the leading breeding companies in using the technology of obtaining F1 diploid hybrids propagated from true seed. However, wide application of this technology is complicated, as germplasm essential for its initiation is either deficient or lacks in international genebanks. Realization of diploid breeding programs is hampered by self-incompatibility and inbreeding depression intrinsic to diploid potatoes, manifested in male sterility or low male fertility and dramatic reduction of productivity in self-pollination progeny. The Institute of Genetics and Cytology of the Academy of Sciences of Belarus was successful in finding solutions to the majority of these problems through long-term research. One of the best collections of diploid breeding material has been created for effective potato breeding via the diploid level, including the production of inbred lines. Objectives. The aim of the study was to evaluate the possibility and specify the peculiarities of using this collection for realizing the program of marker-assisted breeding of diploid potato hybrids, which started in 2021. The objectives of the research were: to evaluate the efficiency of schemes of creating initial materials for establishing diploid inbred lines; to select fertile self-compatible S1(F2) diploid clones assessed by tuber productivity, cultivated potato characters, disease and pest resistance; to produce seeds of S2(F3) selfed clones. Results. It was demonstrated that the schemes most effective for establishing inbred lines were those that involved advanced diploid breeding clones derived from interspecific hybrids and Solanum tuberosum L. fertility and self-compatibility (DFS) donor lines. In total, 3212 seeds (36.5 seeds/berry) were produced from self-pollination of 38 advanced diploid hybrids from five crossing schemes. We failed to produce seeds from self-pollination of backcross progeny of hybrids between dihaploids and DFS to initial dihaploids because of their low male fertility. Backcrossing of such hybrids to DFS demonstrated insufficient level of cultivated potato characters. The self-progeny of 29 diploid clones was selected for establishing S1 selfed clones. The seeds resulting from self-pollination demonstrated low germination and poor seedlings development at the early stage, therefore less than 20% of them were planted in pots for producing 1st year seedlings. Two thirds of seedlings were discarded during harvesting as having failed the assessment of their cultivated potato characters. In total, 167 seedlings of S1 selfed clones from all 29 initial diploid clones were selected. In field trials (Minsk, 2025), 67 out of 156 S1 selfed clones (42.9%) produced berries. From these 67, 36 S1 clones (23.1%) were selected thanks to their cultivated potato characters and field late blight resistance. They originated from 17 initial diploid clones carrying DNA markers of a set of disease and pest resistance genes. The seeds resulting from their self-pollination totaling 4699 pcs, 4382 of which were from S1 clones derived from interspecific hybrids, will be used for the production and selection of S2 self-pollinated clones. Conclusions. The collection of diploid potato used in the research has shown good suitability for the establishment of inbred lines. The most effective schemes for producing the initial material were based on the use of interspecific hybrids and original S. tuberosum fertility and self-compatibility donor lines. Seeds were obtained from self-pollination of fertile self-compatible S1 clones derived from the original diploid clones carrying DNA markers of a number of disease and pest resistance genes of importance for breeding.