Objective: Assess the association between cannabis use and tinnitus in a nationally representative sample of US adults. Study Design: Cross-sectional. Setting: Population-based. Patients: Adults aged 20 to 59 years who participated in 2011 to 2012 and 2015 to 2016 National Health and Nutrition Examination Survey (NHANES) with available data on tinnitus, audiometry, and substance use. Intervention: None.Main Outcome Measure: Tinnitus, demographic information, and medical history were obtained from NHANES questionnaires. Tinnitus was defined as bothersome tinnitus in the past year. Cannabis use was categorized as never use, low-volume use (1-2 pipes/joints per day), and high-volume use (3+ pipes/joints per day). Multivariable regression models with interaction and mediation analyses were conducted. Sampling weights were incorporated to yield results generalizable to the US population. Results: Tinnitus prevalence was significantly higher among high-volume cannabis users (odds ratio [OR], 20.5%; 95% confidence interval [CI], 16.0-26.0%]) and low-volume users (OR, 17.0%; 95% CI, 14.3-20.0%) than nonusers (OR, 12.0%; 95% CI, 10.4-13.9%). High-volume cannabis use was significantly associated with tinnitus relative to nonusers in multivariable models adjusting for demographics, cardiovascular factors, hearing loss, noise exposure, and depression (OR, 2.05; 95% CI, 1.1-3.9). Tinnitus severity was comparable among high volume, low volume, and noncannabis users. There was no significant mediation or interaction of depression affecting the association between cannabis use and tinnitus. Conclusions: Bothersome tinnitus prevalence was significantly higher among cannabis users relative to nonusers. High-volume cannabis use was independently associated with tinnitus in a multivariable model accounting for relevant factors including depression. Future study is warranted to elucidate the impact of various levels of cannabis use on tinnitus.