cua
Scale computer-use agents with open-source drivers, fleets, and benchmarks.
cua is an open-source project that gives computer-use AI agents computers to work with, offering open-source drivers, cross-OS fleets and benchmarks. Recent coverage includes the team's release of CUA-S1-FORMS, a narrow decision model for form interactions with 706k parameters, a 2.8 MB checkpoint and under 30 minutes of training, which predicts whether to use, check, click or skip elements.
Scale computer-use agents with open-source drivers, fleets, and benchmarks.
Open-source drivers, cross-OS fleets, and benchmarks for computer-use AI agents.
Open-source drivers and benchmarks giving AI agents computers they can use
The Cua team released CUA-S1-FORMS, a narrow decision model for form interactions with only 706k parameters and a 2.8 MB checkpoint, trained in under 30 minutes. Inspired by TypeSafe's JEV and Kahneman's dual-system theory, it predicts whether to use, check, click, or skip given elements, without generating new values or considering screenshots.
Open-source desktop automation and cloud desktops giving AI agents computers to use.