Rationale <p> Oxytocin decreases alcohol consumption across a variety of non-human animal models. However, clinical trials using this peptide have produced mixed results. Among the reasons for difficulties in translating preclinical findings of promising treatments for alcohol use disorder to clinical applications is that many non-human animal studies do not sufficiently model the human conditions.</p> Objectives <p> The current study used the translationally relevant intranasal route of administration of peptides and social housing to test the effects of oxytocin (Oxt) and arginine vasopressin (AVP) on voluntary alcohol drinking in mice.</p> Methods <p> All experiments used male and female C57BL/6J mice drinking alcohol or sweetened alcohol and water in a 2-bottle choice procedure. Intranasal Oxt (3 mg/kg) was administered to single-housed mice. Intranasal Oxt (3 mg/kg) or AVP (3 or 1 mg/kg) were administered to group-housed mice. </p> Results <p> Intranasal Oxt decreased alcohol intake in single housed mice, but was ineffective in decreasing drinking of alcohol or sweetened alcohol in group-housed mice. In contrast to Oxt, AVP decreased consumption of sweetened alcohol in group-housed mice. This effect was blocked by AVP receptor 1a antagonist and was non-selective to alcohol at higher (decreased sucrose consumption), but not lower, doses of AVP.</p> Conclusions <p> Our studies for the first time demonstrate that Oxt loses its effectiveness when administered to rodents in translationally relevant conditions (intranasal route and social housing), which could reflect its limited success in clinical trials. In addition, they suggest that intranasal AVP could be more effective than Oxt in decreasing alcohol intake in these conditions.</p>