cua
Open-source infrastructure for Computer-Use Agents. Sandboxes, SDKs, and benchmarks to train and evaluate AI agents that can control full desktops (macOS, Linux, Windows).
Cua is an open-source platform designed for building, benchmarking, and deploying AI agents capable of interacting with any computer. It provides isolated, self-hostable sandboxes using technologies like Docker, QEMU, and Apple Vz for agentic UI automation and secure code execution.
Features
Compatibility
Use cases
Alternatives
Related searches
Comments
- Morgan LeeMay 10, 2026
Good for researchers building and evaluating computer-use AI systems.
- SScout WhiteApr 8, 2026
Sandbox isolation makes it safe to run agents that control a computer.
- PParker HarrisMar 25, 2026
Open-source computer-use agent infrastructure with sandboxes and benchmarks.
- DDylan DavisMar 13, 2026
Benchmark suite enables reproducible comparison of computer-use agent approaches.