The U.S. loneliness epidemic is associated with substantial morbidity and mortality. While higher social media use (SMU) has been associated with higher loneliness among youth, these associations have not been sufficiently examined in adult populations. Additionally, insufficient research has assessed both SMU time and frequency in the same study. Therefore, the primary aim was to evaluate associations between SMU, both by time and frequency, and loneliness in a nationally representative sample of U.S. adults. We recruited 1512 U.S. adults ages 30–70 in 2023. We assessed loneliness using the NIH PROMIS four-item scale and self-reported SMU time and SMU frequency. Survey-weighted logistic regression models determined associations between both SMU measures and loneliness, controlling for gender, age, sexual orientation, educational attainment, employment status, and marital status. Both SMU time and SMU frequency were independently and linearly associated with loneliness (p < 0.001 for both). Although odds of loneliness increased for each increase in frequency, the association between time spent on social media and loneliness demonstrated an inverted U-shape with maximal loneliness in the third quartile of SMU. Results suggest that both time and frequency of SMU may be useful targets for interventions aimed at curbing the negative impact of SMU on loneliness.