Introduction. A large number of studies have been devoted to the study of spillover effects in many sectors of the economy. However, inflation spillovers in regional data from Russia are studied for the first time. Theoretical analysis. This paper aims at studying inflation spillovers in the aggregate and 12 subgroups of the CPI in 26 regions of the Russian Federation. To achieve this goal, we use the Diebold and Yilmaz methodology to quantitatively assess the spillover effect for the consumer price index. Diebold and Yilmaz proposed to calculate a number of indices based on the decomposition of the variance vector autoregressive model to characterize the connectivity of the system at various levels, from pairwise to system-wide. The decomposition of the variance shows what part of the future uncertainty of region i is associated with shocks in region j. The article implements three vector autoregressive models (VAR) with thinning of coefficients (with a penalty) – the Elastic Net, Lasso and Ridge VAR models. Empirical analysis. The dynamic relationship of random shocks (spillover effects) of inflation between regions of the Russian Federation and the stability of the selected methods are studied. To calculate the spillover measures, a technique based on the decomposition of the forecast variance of multivariate time series is used. Clustering methods are used to identify spatial macroregions of the spread of inflation spillovers. Results. The transmission of random shocks on a regional basis during the period from January 2002 to June 2024 occurs from the central regions to the Siberian regions and to the Far Eastern regions. The results of the study provide new information on the heterogeneity of inflation spillovers between Russian regions.