Lifetime relapse rates remain a major obstacle in addressing the current opioid crisis. Relapse to opioid use can be modeled in rodent studies where drug self-administration is followed by a period of abstinence and a subsequent test for drug seeking. Abstinence can be achieved through extinction training, forced abstinence, or voluntary abstinence. Voluntary abstinence can be accomplished by introducing adverse consequences of continued drug self-administration (e.g., punishment or electric barrier) or by introducing an alternative nondrug reward in a discrete choice procedure (drug versus palatable food or social interaction). In this review, we first discuss pharmacological and circuit mechanisms of opioid seeking, as assessed in the classical extinctionreinstatement model, where reinstatement is induced by reexposure to the self-administered drug (drug priming), discrete cues, discriminative cues, drug-associated contexts, different forms of stress, or withdrawal states. Next, we discuss pharmacological and circuit mechanisms of relapse after forced or voluntary abstinence, including the phenomenon of "incubation of heroin craving" (the time-dependent increases in heroin seeking during abstinence). We conclude by discussing future directions of preclinical relapse-related studies using opioid drugs.Neuropsychopharmacology (2019) 44:465-477; https://doi.[143] 1Note: Several reinstatement-related papers published results with more than one reinstating stimulus (e.g., drug priming and stress, drug priming and drug cue) and appear in more than one category. Historical citations refer to the initial papers published with a given model.Relapse to opioid seeking in rat models: behavior, pharmacology and. . . DJ Reiner et al.Neuropsychopharmacology (2019) 44:465 -477 1234567890();,:Relapse to opioid seeking in rat models: behavior, pharmacology and. . . DJ Reiner et al.