Psychopathy Checklist-Revised (PCL-R) is the gold standard for assessing psychopathy. However, its factorial structure has been subject of debate, and different factor models have been proposed. Furthermore, research has not focused on the PCL-R factorial structure among intimate partner violence (IPV) perpetrators. We analyzed and compared the fit of nine alternative models of PCL-R among a sample of 242 IPV perpetrators. Results revealed PCL-R has adequate factor validity. The three-factor model without testlets provided the best fit to the current data. The three-factor model also showed good reliability, and the factors established differential associations to penal and personal variables thus revealing discriminant validity. In the current study, psychopathy was thus best conceptualized as a clinical construct comprising a particular affective functioning (e.g., shallow affect, lack of empathy), and a specific interpersonal (e.g., egocentricity, deception) and behavioral style (e.g., irresponsibility, impulsivity) among IPV perpetrators.