cua
Open-source infrastructure for training and evaluating AI agents to control full desktop environments.
Pricing
Free tier
Flat rate
Adoption
↗RisingLicense
Open Source
Data freshness
Verified · Jul 16, 2026Overview
What is cua?
Cua provides sandboxes, SDKs, and benchmarks for developing Computer-Use Agents that can operate across macOS, Linux, and Windows. It's essential for researchers and developers working on AI-driven automation in desktop environments.
Key differentiator
“Cua stands out by offering a comprehensive open-source solution for training and evaluating AI agents across multiple desktop operating systems, providing unique cross-platform capabilities not found in other tools.”
Capability profile
Capability Radar
Honest assessment
Strengths & Weaknesses
↑ Strengths
↓ Weaknesses
API requires Python-specific patterns, TypeScript SDK is community-maintained
v0.1 to v0.2 migration required rewriting chain definitions
Advanced features lack detailed examples and explanations in the official docs
Known performance degradation when managing a high number of concurrent Computer-Use Agents
Fit analysis
Who is it for?
✓ Best for
Researchers looking to evaluate the performance of AI agents in controlling desktop environments across multiple operating systems.
Developers building automation tools that need to interact with a wide range of desktop applications and interfaces.
✕ Not a fit for
Projects requiring real-time interaction or streaming capabilities, as cua focuses on batch processing and evaluation.
Teams looking for cloud-based solutions, as it is designed for self-hosted environments.
Cost structure
Pricing
Free Tier
Available
Open source — free to use
Starts at
$0
Model
Flat rate
Enterprise
None
Performance benchmarks
How Fast Is It?
Ecosystem
Relationships
Works well with
Integrations
Next step
Get Started with cua
Step-by-step setup guide with code examples and common gotchas.