Purpose of Review <p>Cannabis Use Disorder (CUD) is increasingly prevalent with the expanding legalization of cannabis, yet no FDA-approved treatments are currently available. Preclinical models have been critical for advancing understanding of the neurobiology of cannabis use and for identifying potential therapeutic targets for CUD. However, progress in models of cannabis self-administration has lagged behind other areas, due to inherent pharmacological and methodological challenges. This review provides a comprehensive evaluation of various preclinical models used to study CUD, including recent innovations in self-administration approaches, with an emphasis on their contributions to medication development.</p> Recent Findings <p>Recent advances in preclinical models, particularly in cannabinoid self-administration procedures, have improved both reliability and translational relevance. New technologies and routes of administration, such as drug vaporization, have been developed to better model human use. Recent studies have provided new insights into the neurocircuitry underlying CUD, highlighting novel opportunities for pharmacological intervention. Emerging targets include novel modulators of the cannabinoid-1 receptor, endocannabinoids, and corticostriatal neurotransmission.</p> Summary <p>Despite challenges and limitations inherent in current preclinical models, they remain indispensable tools in advancing understanding of CUD and facilitating the development of novel treatments. Continued innovation in model design, combined with greater integration of preclinical and clinical research, is essential to bridge the gap between basic science and the clinical application of new therapies for CUD.</p>