Using a sample of 1227 Swedish adolescents we examined peer relations across contexts and for multiple peer targets (three ''Very Important Persons'', VIPs). Specifically, we examined the relations between antisocial behaviour and the types of relationships individuals had with their VIPs (e.g. friend, romantic partner), the contexts in which they had met, and where they spent time (e.g. school, neighbourhood, club). Additionally, we tested an ''additive homophily'' hypothesis, or the idea that youths would show unique similarities to multiple peers. Results showed that individuals who nominated romantic partners as their first VIPs were the most antisocial (both boys and girls), and individuals who had met and spent time with their first VIPs in the neighbourhood were also the most antisocial. Similar results were found for the antisocial behaviour of the first VIP. Finally, results supported the additive homophily hypothesis, showing that significantly more variance in individual behaviour is explained when including second and third VIPs.