The pollution of indoor air is a major worldwide concern in our modern society for people's comfort, health, and safety. In particular, toluene, present in many substances including paints, thinners, candles, leathers, cosmetics, inks, and glues, affects the human health even at very low concentrations throughout its action on the central nervous system. Its prevalence in many workplace environments can fluctuate considerably, which led to firm regulation with exposure limits varying between 50 and 400 ppm depending on exposure time. This therefore requires the development of technologies for an accurate detection of this contaminant. Metal−organic frameworks have been proposed as promising candidates to detect and monitor a series of molecules at even extremely low concentrations owing to the high tunability of their functionality. Herein, a high-throughput Monte Carlo screening approach was devised to identify the best MOFs from the computation-ready, experimental (CoRE) metal−organic framework (MOF) density-derived electrostatic and chemical (DDEC) database for the selective capture of toluene from air at room temperature, with the consideration of a ternary mixture composed of extremely low-level concentration of toluene (10 ppm) in oxygen and nitrogen to mimic the composition of air. An aluminum MOF, DUT-4, with channel-like micropores was identified as an excellent candidate for the selective adsorption of toluene from air with a predicted adsorption uptake of 0.5 g/g at 10 ppm concentration and room temperature. The toluene adsorption behavior of DUT-4 at low equivalent concentrations, alongside its sensing performance, was further experimentally investigated by its incorporation in a quartz crystal microbalance sensor, confirming the promises of DUT-4. Decisively, the resulting high sensitivity and fast kinetics of our developed sensor highlight the applicability of this hand-in-hand computational−experimental methodology to porous material screening for sensing applications.