Abstract Background: Genomic sequencing data from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have provided important information on the emergence of variants of interest. Some research has focused on determining intra-host variability and evolution, and others have focused on viral genetic variability in different tissues, as well as implications for virus infectivity and pathogenicity. However, the inter-host variability of SARS-CoV-2 in respiratory compartments and its implications has not yet been explained. To determine a possible pattern of tissue-specific inter-host variability at different levels of the respiratory system associating this information with clinically important characteristics. Methods: In this work, we analyzed the inter-host variability of SARS-CoV-2 using the Shannon entropy method with genomic data from NCBI database. Fifty-seven SARS-CoV-2 viral genomic sequences were included in the study. Oronasopharyngeal (n = 30) and tracheal aspirate (n = 27) sample sequences were analyzed by bioinformatics methods to determine the viral variability. Results: Different patterns of variability were observed in SARS-CoV-2 in the respiratory tract, which may indicate tissue compartmentalization phenomena. Genetic changes found in the N y S gene could affect the detection and spread of the virus. The overall analysis of the viral genomes identifies the variants of worldwide distribution. The viral diversity found in the different parts of the respiratory tract was high and uniformly distributed throughout the genome. Conclusions: Tracking inter-host variations in SARS-CoV-2 could help to understand the effects of viral evolution, provide information for diagnosis, drug, and vaccine design.