Online hate speech is a matter of concern for social media platforms, regulators, researchers, and the public. Despite its widespread prevalence and contentious nature, little research has been done on the perception of hate speech and its psychosocial predictors. To address this gap, we conducted a study on the perception of hate speech toward migrants in online comments, analyzing the differences between a public group (NPublic = 649) and an expert group (NExperts = 27) and exploring the correlation between the proposed hate speech indicators and perceived hate speech in both groups. Additionally, we explored various predictors of hate speech perception, including demographic and psychological variables such as human values, prejudice, aggression, impulsiveness, social media behavior, attitudes toward migrants and migration, and trust in institutions. Our results show that the public and experts have differing sensitivities toward hate speech, with the expert group perceiving comments as more hateful and emotionally harmful compared with the general population, who tend to agree more with antimigrant hateful comments. The proposed hate speech indicators and especially their total scores have a strong correlation with both groups' perceptions of hate speech. Psychological predictors, such as the human values of universalism, tradition, security, and subjective social distance, were significant predictors of online hate speech sensitivity. Our findings emphasize the need for public and scholarly discussions, more robust educational policies, and intervention programs with specific measures to counter hate speech online.