To test the efficacy of an intervention to reduce tobacco use among youth (10-19 years old) in slum communities in Delhi, India. This community-based cluster-randomized trial included 14 slums composed of purposely built resettlement colonies and adjacent inhabitant-built Jhuggi Jhopris. Youth in the intervention received a 2 year multiple-component intervention: (a) youth and adult leader training; (b) peer-led interactive activities and outreach; (c) tobacco cessation camps; and (d) enforcement of India's Tobacco Control Law (smoke-free environments and youth access). Overall, no differences between the intervention and control conditions were observed over time; self-reported tobacco use declined in both groups. However, when stratified by type of residence, a significant decrease was observed among youth in the resettlement colonies in the intervention group for overall tobacco use (slope = -0.69) and cigarette and bidi smoking (slope = -0.66), compared to an increase in the control group (slope = 0.24 and 0.12, respectively) (P < 0.001). No differences in smokeless tobacco (SLT) use were observed for either group. Comprehensive community-based interventions that engage youth can be effective in reducing smoking among disadvantaged youth in India. More intensive interventions, like tax increases or large-scale media campaigns, appear warranted for the most marginalized in this context and for SLT products.