The incidence of ischemic stroke remains high in many countries, despite a declining trend in the incidence of brain circulation disorders. Limited knowledge of the pathogenesis, diagnosis, clinical presentation, and treatment of this life-threatening disease can be complemented by modelling of ischemic stroke on animals, particularly, in rodents (rats, mice). The aim of review: to provide a rationale for choosing an optimal model of human ischemic stroke. Among more than 300 primary literature sources from various databases (Scopus, Web of science, RSCI, etc.), 84 sources were selected for evaluation, of which 72 were published in the recent years (2015–2019). The criteria for exclusion of sources were low relevance and outdated data. The review examined six experimental models of acute cerebrovascular events including middle cerebral artery occlusion, embolic stroke model; phototrombosis; models using thrombin, endothelin-1 and electro-coagulation of the middle cerebral artery. The review outlines the factors influencing reliability and reproducibility of research results, related to adherence to the rules of animal accommodation and acclimatization, nutrition and care, selection of anesthesia and pain relief methods, compliance with aseptic techniques, monitoring of basic physiological parameters at all stages of the ischemic stroke modeling experiment, with humane withdrawal of animals from the experiment. It was concluded that the model of ischemic stroke, based on the occlusion of the middle cerebral artery, is most promising as being the closest to features of human ischemic stroke and enabling to obtain reproducible results in the experiment. © 2020, V.A. Negovsky Research Institute of General Reanimatology. All rights reserved.