Cognitive impairments are common in homeless youth and negatively impact academic and vocational outcomes. We examined the feasibility and efficacy of cognitive interventions provided to 18- to 22-year-old homeless youth living in urban supportive housing. Ninety-one homeless youth were randomized to receive either targeted cognitive training (cognitive remediation) or general cognitive activation (computer skills training). Cognitive and psychological outcomes were assessed at baseline, after 13 and 26 sessions, and 1 month postintervention. A high dropout rate highlighted the feasibility challenges of treating this population. Intent-to-treat analysis found significant improvements across groups in specific and global measures of cognition and psychological distress, with no significant group differences. Transition-age homeless youth show improvements in cognitive and psychological functioning when engaged in interventions that address their cognitive development. This speaks to the malleability of cognitive skills in this cohort and lays the groundwork for future research to address their cognitive health.