Objectives: This study aimed to investigate tobacco use behaviours and their correlates among secondary school students in Nepal and Sri Lanka together with cross-country comparisons. Design: Cross-sectional survey. Methods and Settings: The data were obtained from the Global Youth Tobacco Survey (GYTS), 2007. Current tobacco use was considered as a response variable. Predictors were selected based on existing literature and theories on adolescent tobacco use. The data of 1,444 Nepalese and 1,377 Sri Lankan students aged 13-15 years was used for analysis. Results: Prevalence of tobacco use varies with 9.4% and 9.1% among Nepalese and Sri Lankan students respectively. Boys (13.2 vs. 5.3 for Nepal, 12.4 vs. 5.8 for Sri Lanka), older and senior students was more likely to be tobacco users in both countries. The average age of tobacco initiation was 10.2 years in Nepal and 8.6 years in Sri Lanka. Factors, namely, individual characteristics, tobacco use among friends, smoking at home and public places, free tobacco products, and lessons about negative effects of tobacco in class were significantly associated with tobacco use in Nepal and Sri Lanka. For instance, friends' tobacco use was a good predictor for adolescents' usage (odds ratio [OR] = 4.0, confidence interval [CI] = 2.61-6.23 in Nepal; OR = 2.5, CI = 1.50-4.23 in Sri-Lanka). Similarly, course curriculum significantly reduced smoking among students. Conclusion: Prevention of tobacco use among school students should be top priority of a country as they are long-term customers and replacement smokers who quit or die. Therefore comprehensive strategies as we proposed along with existing prevention programmes should be tightened to stop them from hazardous behaviours.