In the US, men who have sex with men (MSM) represent the largest proportion of AIDS deaths, people living with AIDS, and HIV incidence. While numerous prevention efforts targeting MSM have successfully reduced HIV incidence in the US, many men are not reached, including gay couples. Several epidemiological studies attribute high rates of new HIV infections to condomless anal sex within male primary partnerships, but few prevention interventions are available to address and reverse this trend. Several factors have recently changed the HIV prevention landscape, including advancements in biomedical strategies, an increased desire to develop interventions that can be scaled up efficiently and a need to develop cost-effective programing due to decreases in prevention dollars. Multi-session individual- and group-level in-person sexual risk reduction interventions for MSM were the gold standard for many years. Behavioral interventions have also been delivered via the internet, since they are convenient, able to reach large numbers of individuals and less labor intensive to implement, compared to interventions delivered via multiple in- person sessions. It is important to examine the optimal modes of providing prevention interventions in the context of this changing prevention landscape. The majority of prevention interventions targeting MSM are individual-focused. However, over the years, our research has identified unique prevention needs for gay couples that have not been sufficiently addressed in individual-focused interventions. For example, while most gay couples have agreements about sex with partners outside the relationship, many couples have difficulty negotiating their agreements, some have difficulty adhering to their agreements and those who break their agreements have difficulty disclosing the break to their partner; all of which could pose potential HIV risk for the couple. Our study team has led the field in examining relationship dynamics among gay couples. With a goal to develop a theory driven yet empirically-informed intervention, we have been studying the nuances of relationships and determining the HIV risk factors that either persist or change over time. Several years of longitudinal quantitative and qualitative studies, have yielded the evidence for an empirically- driven model that provides a foundational framework for the proposed intervention. Additionally, due to a dearth of prevention interventions for gay couples, there is little documented knowledge about the best modes of intervention delivery. Specifically, we do not know if it is efficacious for couples to participate in an intervention in-person or online. We also do not know if particular topics are best addressed using one of these modes or the other. To address these questions, we propose in an R01 application, to test the efficacy of our theory-based and empirically-driven HIV risk reduction intervention for gay male couples using a randomized controlled trial.