Acceptance and Commitment Therapy (ACT) is a contextual cognitive-behavioural approach with a developing evidence base for clinical and cost-effectiveness as an individually-delivered intervention to promote recovery from psychosis. ACT also lends itself to brief group delivery, potentially increasing access to therapy without inflating costs. This study examined, for the first time, the feasibility and acceptability of ACT groups for people with psychosis (G-ACTp).Participants were recruited from community psychosis teams. Ratings of user satisfaction, and pre-post change in self-rated functioning (primary outcome), mood (secondary outcome) and ACT processes were all completed with an independent assessor. Of 89 people recruited, 83 completed pre measures, 69 started the four-week G-ACTp intervention, and 65 completed post measures.Independently assessed acceptability and satisfaction were high. Functioning (Coeff. = -2.4, z = -2.9, p = 0.004; 95% CI: -4.0 to -0.8; within subject effect size (ES) d = 0.4) and mood (Coeff. = -2.3, z = -3.5, p = 0.001; 95% CI: -3.5 to -1.0; d = 0.4) improved from baseline to follow-up. Commensurate changes in targeted ACT processes were consistent with the underlying model.The uncontrolled, pre-post design precluded blinded assessments, and may have inflated effect sizes. Participants may have improved as a result of other factors, and findings require replication in a randomized controlled trial (RCT).This preliminary study showed that brief group ACT interventions for people with psychosis are feasible and acceptable. Uncontrolled, pre-post assessments suggest small clinical improvements, and changes in psychological processes consistent with an ACT model. Replication in an RCT is required, before implementation can be recommended.