A team of researchers from UC Berkeley have released CUA-Lite, an open platform for computer-use agents (CUAs). The argument behind it is infrastructural rather than model-centric: training and benchmarking a CUA requires four pieces: agents, environments, traces, and a framework to evaluate and train them and all four are currently fragmented across separate repositories with incompatible interfaces. CUA-Lite puts them behind one action space, one data schema, and one command, across desktop, browser and mobile.
Is it deployable? Yes. The stack installs with uv sync --all-extras on Python 3.12, and its lightweight sandboxes run on any Docker host without /dev/kvm, so cloud instances, CI runners and nested containers all work.
The VM tax, and how Lite.OSWorld removes it
The most concrete contribution is Lite.OSWorld. OSWorld provides a faithful Ubuntu desktop, but it ships as a full QEMU/KVM virtual machine per task, requiring nested virtualization that most managed infrastructure does not expose. CUA-Lite reproduces the same task suite and the same evaluators on a GNOME desktop inside a plain Docker container.
TaskOSWorldLite.OSWorldRuntimeQEMU/KVM VMDocker containerHost requirement/dev/kvm, nested virtAny Docker hostMemory4.1 GB0.9 GBCold start29.9 s23.8 sParallelismbaseline~4.6× more instancesTask suiteOSWorldIdentical
