Cua logo

Cua

Cua AI Agent
Rating:
Rate it!

Overview

Open-source computer-use tooling for desktop automation, cloud desktops, local macOS VMs, decision models, and agent benchmarks.

Cua is an open-source computer-use project for giving agents usable desktop environments and automation tools. The GitHub project describes support for desktop automation, isolated cloud desktops, local macOS VMs, specialist decision models, and benchmarks for evaluating computer-use agents. It is aimed at builders who want to bring their own agent and model, operate graphical applications, provision Linux desktops, run local macOS VMs, and evaluate agent behavior on simulated computer-use tasks.

AI Agent Store research

What the evidence says about Cua

Cua is distinctive as an open-source computer-use stack focused on giving agents desktop environments they can operate, rather than only exposing APIs or text tools. The repository positions it around open-source drivers, isolated cloud desktops, local macOS VMs, specialist decision models, and benchmarks for training, evaluation, and data generation.

Verified September 22, 2026

Verified capabilities

  • Desktop automation for agents

    The project describes Cua as providing open-source desktop automation and computer-use tooling that lets agents use graphical desktop environments.[1]

  • Isolated cloud desktops

    Cua includes Cua Fleets, described in the repository excerpt as a way to provision a Linux desktop, run a command, and save a screenshot.[1]

  • Local macOS VMs

    The README excerpt lists local macOS VMs and includes Lume as a path for creating a Tahoe VM and connecting over SSH.[1]

  • Specialist decision models

    The project references CUA-S1 as small, specialized models for computer-use decisions.[1]

  • Benchmarks for evaluation

    Cua Bench is described as supporting simulated task creation and verification for evaluating computer-use agents.[1]

  • Bring-your-own agent and model

    The official excerpt states that users can bring their own agent and model, with Cua providing the computer and automation tools.[1]

Where it fits best

  • Developers and research teams building or evaluating computer-use agents that need desktop automation, provisioned desktops, VM environments, or benchmark tasks.[1]
  • Teams that want to bring their own agent and model while using Cua to provide the computer environment and automation tools.[1]

Buying and deployment notes

The supplied official context shows a public GitHub repository described as open-source and displaying an MIT license. No separate hosted pricing details are provided in the supplied context.[1]

Platforms: GitHub, Linux desktop, macOS VM[1]

Deployment: open-source repository, local VM, cloud desktop[1]

Important considerations
  • The supplied context verifies the open-source GitHub project but does not provide detailed paid pricing, hosted-service limits, or commercial plan information for Cua Fleets.[1]
  • Cua is presented as infrastructure and tooling for computer-use agents, including an option to bring your own agent and model, so buyers looking for a fully packaged end-user assistant should evaluate fit carefully.[1]
Sources and research method (1)

We record only claims tied to public sources checked by our team or listing workflow. Counts above are derived directly from this profile, not a subjective rating.

  1. GitHub - trycua/cua: Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.GitHub · checked 2026-09-22

Autonomy level

50%

Comparisons


Custom Comparisons

Some of the use cases of Cua:

  • building computer-use agents
  • desktop automation experiments
  • agent evaluation and benchmarking
  • provisioning isolated desktops for agent workflows
  • testing GUI interaction tasks

Loading Community Opinions...

Pricing model:

Code access:

Popularity level: %

Free credibility widget

Turn this profile into a trust signal

Show prospects that Cua has a public place where they can check product details, pricing, ratings, and reviews.

Build confidence

Give buyers a third-party profile to explore.

Reduce hesitation

Put validation beside your strongest CTA.

Earn discovery

Every badge links prospects to your listing.

Choose your style

Preview it, then copy the complete embed code.

Live previewReady to embed

Shows buyers where to validate your product, pricing, and reputation.

Plain HTML. No signup, script, or maintenance required.

Make it work for you

Describe the job. Get an AI worker you can actually message.

We create the setup, keep it running after your laptop closes, and save its memory. Test in the browser, then add Telegram, WhatsApp, or Slack.

Runs without your laptopBrowser + messaging appsCredits, keys, or subscriptionsMemory survives restarts

Plans start at $29/month. Cancel anytime.

Hosted agent

OpenClaw or Hermes

saved state
Browser
WhatsApp
Telegram
Slack
“I checked the inbox, handled the routine messages, and sent you the one question that needs a decision.”
Create an AI worker that keeps running after this tab closes.
Open Agent Teams

Did you find this page useful?

Not useful
Could be better
Neutral
Useful
Loved it!