Given the many instances of violence and crime that occur as a consequence of psychopathy, it is vital to find those variables that can reduce the expression of such behaviors. In this regard, one potentially useful variable is that known as Emotional Intelligence (EI) or the ability to perceive, use, understand, and regulate emotions. EI has been categorized according to three main approaches: performance-based ability, self-report ability, and self-report mixed models. Given the emotional deficits of the psychopathic population, EI could be a protective factor. Several studies have analyzed the relationship between EI and psychopathy, but the results are unclear. This disparity may be due to the EI model employed to measure EI. The aim of our research is to systematically review the relationship between the different models of EI and psychopathy, both in the total and clinical/inmate sample. We searched Scopus, Pubmed, and PsicINFO to find relevant articles. Twenty-nine eligible studies were found. They were divided according to the model of EI and the sample used. The results for the total sample differ according to the measure of EI: when the performance-based ability model is used, the majority of studies find a negative relationship between EI and psychopathy. When using self-reports, the results are inconsistent. The findings with the clinical/inmate sample are in the same direction as the total sample. In conclusion, the results suggest that higher EI abilities measured through performance-based ability models—but not through self-reports—are related to lower psychopathy deficits. Limitations and clinical implications are discussed.