Past research alerts to the increasingly unpleasant climate surrounding public debate on social media. Female politicians, in particular, are reporting serious attacks targeted at them. Yet, research offers inconclusive insights regarding the gender gap in online incivility. This paper aims to address this gap by comparing politicians with varying levels of prominence and public status in different institutional contexts. Using a machine learning approach for analyzing over 23 million tweets addressed to politicians in Germany, Spain, the United Kingdom, and the United States, we find little consistent evidence of a gender gap in the proportion of incivility. However, more prominent politicians are considerably and consistently more likely than others to receive uncivil attacks. While prominence influences US male and female politicians’ probability to receive uncivil tweets the same way, women in our European sample receive incivility regardless of their status. Most importantly, the incivility varies in quality and across contexts, with women, especially in more plurality contexts, receiving more identity-based attacks than other politicians.