AgentIndex icon
AgentIndex
cua logo
cua
★ 31.1k
vs
on-policy logo
on-policy
★ 2.1k

cua vs on-policy

cua: 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.; on-policy: This repository implements MAPPO, a multi-agent variant of PPO, widely used in cooperative multi-agent games and research. It provides robust implementations for various multi-agent environments like StarCraft II, Hanabi, and Google Research Football, along with detailed training scripts and hyperparameter guidance.

01

TL;DR

cua logo
cuaLLM Infra · ★ 31.1k
When to choose:
  • You are building Computer-Use agents and need robust, self-hostable isolated desktop/VM sandboxes (Docker, QEMU, Apple Vz).
  • You need native app operating drivers, standardized benchmarking harnesses, and secure isolation for autonomous agent task execution.
When to skip:
  • You only need simple UI macros on your personal physical machine (Windows-MCP or local desktop helpers are easier).
  • Your agents interact solely via text APIs, shell commands, or databases and have no need for a virtual graphical desktop.
Read cua full review →
on-policy logo
on-policyLLM Infra · ★ 2.1k

Research and experimentation in cooperative multi-agent reinforcement learning

Read on-policy full review →
02

Side-by-Side Comparison

Field
cua logocua
on-policy logoon-policy
Category
LLM Infra
LLM Infra
Stars
★ 31.1k
★ 2.1k
License
MIT
MIT
Install Command
curl -fsSL https://cua.ai/install.sh | sh
pip install -e .
Updated
1d ago
2y ago
Open Source
Yes
Yes
Website
↗ Visit
↗ Visit
GitHub
↗ GitHub
↗ GitHub
Tags
AI Agents, Virtualization, UI Automation
Multi-Agent Reinforcement Learning, PPO, MAPPO
03

Features

cua logocua
01Build AI agents for desktop UI automation and interaction.
02Provide isolated code execution environments (sandboxes).
03Benchmark computer-use models with standardized tasks.
04Train agents using reinforcement learning environments.
05Manage high-performance macOS/Linux VMs on Apple Silicon.
on-policy logoon-policy
01Implementation of MAPPO (Multi-Agent PPO)
02Support for diverse multi-agent environments (e.g., StarCraft II, Hanabi)
03Ready-to-use training scripts for various scenarios
04Detailed hyperparameter guidance and updated results
05Default support for shared policy among agents
04

Use Cases

cua logocua
↳Developing and deploying AI agents for autonomous desktop interaction and task completion.
↳Creating secure, isolated code execution environments for AI coding assistants and development workflows.
↳Benchmarking and training computer-use agents using standardized tasks and reinforcement learning.
on-policy logoon-policy
↳Research and experimentation in cooperative multi-agent reinforcement learning
↳Benchmarking and evaluating PPO's effectiveness in MARL scenarios
↳Training AI agents for popular multi-agent games like StarCraft II and Hanabi
05

Best For

cua logocua
Most PopularEssentialTrending
on-policy logoon-policy
—
FAQ

FAQ

What is the difference between cua and on-policy?
Both cua and on-policy are in the LLM Infra category. cua has 31.1k stars, while on-policy has 2.1k stars.
Which is better, cua or on-policy?
The best choice depends on your use case. Choose cua if Developing and deploying AI agents for autonomous desktop interaction and task completion., and on-policy if Research and experimentation in cooperative multi-agent reinforcement learning.
Is cua free or open source?
Yes, cua is open source on GitHub (MIT).
Is on-policy free or open source?
Yes, on-policy is open source on GitHub (MIT).
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
→

Related

Alternatives to cua →Alternatives to on-policy →cua details →on-policy details →cua vs mcp-server-browserbase →