The COVID-19 pandemic has underlined the need for effective models to evaluate potential therapeutic options in clinical settings. While animal models are crucial for understanding host-pathogen interactions and disease mechanisms, there is a lack of methods to correlate experimental results with human data. High-throughput techniques like single-cell transcriptomics offer detailed insights into molecular and cellular changes during inflammation, but computational tools to reliably compare interspecies differences in model organisms are still limited. Here, we present a neural network-based framework to map temporal disease states of two hamster models ? Syrian hamsters, which exhibit moderate disease, and Roborovski hamsters, which experience severe disease outcomes following infection ? to human COVID-19 severity levels. By quantifying the overall transcriptomic similarity of individual cell types across species and severity levels, we found that the profiles of most Syrian hamster cell types aligned closely with those of patients experiencing moderate COVID-19, while Roborovski hamsters shared the highest similarities in neutrophils with that of severe COVID-19 patients. Transcriptome-wide analysis and candidate gene expression further revealed similarities between hamster and human immune responses, particularly involving monocytes and neutrophils. Disease-linked pathways found in all species specifically related to interferon response or inhibition of viral replication. We postulate that our structured neural network-supported workflow could be applied to other diseases, allowing better identification of suitable models with similar pathomechanisms.