Baclofen and γ-hydroxybutyrate (GHB) exert γ-aminobutyric acid (GABA)B receptor agonism and have therapeutic utility but possess different pharmacological activities. We examined whether separate groups of mice could be trained to discriminate either baclofen or GHB, and the contribution of GABAB receptors to discriminative stimulus effects. Male C57BL/6J mice were trained to discriminate either baclofen (3.2 mg/kg, intraperitoneal) or GHB (178 mg/kg, intraperitoneal) from saline under a fixed-ratio 10 schedule. The GABAB antagonist 3-aminopropyl(diethoxymethyl)phosphinic acid (CGP 35348) was used to pharmacologically assess GABAB receptor involvement. The selectivity of the resulting discriminations was assessed with the opioid agonist morphine and the benzodiazepine midazolam. In baclofen-trained mice, both baclofen and GHB were readily discriminated. Baclofen produced a maximum of 86% baclofen-appropriate responding. CGP 35348 (320 mg/kg, i.p.) produced a 4.7-fold rightward shift in the dose-effect function. GHB produced a maximum of 85.8% baclofen-appropriate responding. In GHB-trained mice, both GHB and baclofen were readily discriminated. In GHB-trained mice, GHB produced a maximum of 85.3% drug-appropriate responding; CGP 35348 (320 mg/kg, i.p.) produced a 1.8-fold rightward shift in the GHB discrimination dose-effect function. Baclofen produced up to 70.0% GHB-appropriate responding. CGP 35348 (320 mg/kg, i.p.) significantly antagonized baclofen discrimination and baclofen produced up to 37% GHB-appropriate responding up to doses that disrupted operant responding. Morphine did not produce substitution for either baclofen or GHB. Midazolam produced partial substitution for both. GHB and baclofen discrimination assays in mice provide a useful approach for examining different receptor types mediating the effects of these two drugs.