While group differences in violence have long been a key focus of sociological inquiry, we know comparatively little about the trends in criminal violence for whites, blacks, and Hispanics in recent decades. Combining geocoded death records with multiple data sources to capture the socioeconomic, demographic, and legal context of 131 of the largest metropolitan areas in the United States, this article examines the trends in racial/ethnic inequality in homicide rates since 1990. In addition to exploring long-established explanations (e.g., disadvantage), we also investigate how three of the most significant societal changes over the past 20 years, namely, rapid immigration, mass incarceration, and rising wealth inequality affect racial/ethnic homicide gaps. Across all three comparisons—white-black, white-Hispanic, and black-Hispanic—we find considerable convergence in homicide rates over the past two decades. Consistent with expectations, structural disadvantage is one of the strongest predictors of levels and changes in racial/ethnic violence disparities. In contrast to predictions based on strain theory, racial/ethnic wealth inequality has not increased disparities in homicide. Immigration, on the other hand, appears to be associated with declining white-black homicide differences. Consistent with an incapacitation/deterrence perspective, greater racial/ethnic incarceration disparities are associated with smaller racial/ethnic gaps in homicide.