Background: The extent of the impact and the interconnections among factors within social and residential contexts during the COVID-19 lockdowns on mental well-being remain to be elucidated. We identified latent classes of each of social and residential context during the lockdown and examined their associations with mental well-being among adolescents in France 1-year after the first lockdown. Methods: We used data collected in 2021 in a cross-sectional school-based pilot study for EXIST, from 387 participants ages 12-15 years. Participants reported retrospectively on characteristics of their social and residential contexts during the lockdown, and their current mental well-being in self-report questionnaires. We used latent class analysis to identify latent classes of social and residential contexts, and linear regression models to examine the associations between these contexts and mental well-being. Results: Four social context classes were identified: class 1 "Low opportunity for social contact at home," class 2 "Moderate opportunity for social contact at home," class 3 "High opportunity for social contact at home," and class 4 "Very high opportunity for social contact at home." Relative to class 4, lower levels of mental well-being were observed among adolescents in class 1 (b = -4.08, 95% CI [-8.06; -0.10]) 1 year after the lockdown. We identified four residential context classes based on proximity to nature, type of residence (e.g., apartment, house), and level of neighborhood deprivation. No association was detected between residential context during the lockdown and adolescent mental well-being one-year later. Conclusion: A limited social context may negatively impact adolescent mental well-being during crises.