In order to develop improved anxiolytic drugs, 8-substituted analogs of triazolam were synthesized in an effort to discover compounds with selectivity for α2/α3 subunit-containing GABAA subtypes. Two compounds in this series, XLi-JY-DMH (6-(2-chlorophenyl)-8-ethynyl-1-methyl-4H-benzo [f][1,2,4]triazolo[4,3-a][1,4]diazepine) and SH-TRI-108 [(E)-8-ethynyl-1-methyl-6-(pyridin-2-yl)-4H-benzo [f][1,2,4]triazolo[4,3-a][1,4]diazepine], were evaluated for in vitro and in vivo properties associated with GABAA subtype-selective ligands. In radioligand binding assays conducted in transfected HEK cells containing rat αXβ3γ2 subtypes (X = 1,2,3,5), no evidence of selectivity was obtained, although differences in potency relative to triazolam were observed overall (triazolam > XLi-JY-DMH > SH-TRI-108). In studies with rat αXβ3γ2 subtypes (X = 1,2,3,5) using patch-clamp electrophysiology, no differences in maximal potentiation of GABA-mediated Cl− current was obtained across subtypes for any compound. However, SH-TRI-108 demonstrated a 25-fold difference in functional potency between α1β3γ2 vs. α2β3γ2 subtypes. We evaluated the extent to which this potency difference translated into behavioral pharmacological differences in monkeys. In a rhesus monkey conflict model of anxiolytic-like effects, triazolam, XLi-JY-DMH, and SH-TR-108 increased rates of responding attenuated by shock (anti-conflict effect) but also attenuated non-suppressed responding. In a squirrel monkey observation procedure, both analogs engendered a profile of sedative-motor effects similar to that reported previously for triazolam. In molecular docking studies, we found that the interactions of the 8-ethynyl triazolobenzodiazepines with the C-loop of the α1GABAA site was stronger than that of imidazodiazepines XHe-II-053 and HZ-166, which may account for the non-sedating yet anxiolytic profile of these latter compounds when evaluated in previous studies.