cua (pronounced “koo-ah”) is a high-performance OS virtualization and control framework for AI agents, offering lightweight, near-native-speed containers on macOS and Linux to automate tasks like application control, web browsing, and code execution. The project includes built-in interfaces to major LLM providers and delivers Docker images, a Lume CLI, and Python/Swift SDKs for seamless Apple Silicon VM deployment under the MIT license, enabling rapid development of AI agent solutions.
Source code: https://github.com/trycua/cua
Get started with a Computer-Use Agent UI and a VM with a single command:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/trycua/cua/main/scripts/playground.sh)"
This script will:
- Install Lume CLI for VM management (if needed)
- Pull the latest macOS CUA image (if needed)
- Set up Python environment and install/update required packages
- Launch the Computer-Use Agent UI
Supported Agent Loops
- UITARS-1.5 – Run locally on Apple Silicon with MLX, or use cloud providers
- OpenAI CUA – Use OpenAI’s Computer-Use Preview model
- Anthropic CUA – Use Anthropic’s Computer-Use capabilities
- OmniParser-v2.0 – Control UI with Set-of-Marks prompting using any vision model
- Mac with Apple Silicon (M1/M2/M3/M4 series)
- macOS 15 (Sequoia) or newer
- Disk space for VM images (30GB+ recommended)
Libre Depot original article,Publisher:Libre Depot，Please indicate the source when reprinting:https://www.libredepot.top/5537.html
