trycua/cua

We’re hosting the Computer-Use Agents SOTA Challenge at Hack the North and online!

Track A (On-site @ UWaterloo): Reserved for participants accepted to Hack the North. 🏆 Prize: YC interview guaranteed.
Track B (Remote): Open to everyone worldwide. 🏆 Prize: Cash award.

👉 Sign up here: trycua.com/hackathon

cua (“koo-ah”) is Docker for Computer-Use Agents - it enables AI agents to control full operating systems in virtual containers and deploy them locally or to the cloud.

With the Computer SDK, you can:

• automate Windows, Linux, and macOS VMs with a consistent, pyautogui-like API
• create & manage VMs locally or using cua cloud

With the Agent SDK, you can:

• run computer-use models with a consistent schema
• benchmark on OSWorld-Verified, SheetBench-V2, and more with a single line of code using HUD (Notebook)
• combine UI grounding models with any LLM using composed agents
• use new UI agent models and UI grounding models from the Model Zoo below with just a model string (e.g., ComputerAgent(model="openai/computer-use-preview"))
• use API or local inference by changing a prefix (e.g., openai/, openrouter/, ollama/, huggingface-local/, mlx/, etc.)

CUA Model Zoo 🐨

• human/human → Human-in-the-Loop

Missing a model? Raise a feature request or contribute!

Quick Start

• Get started with a Computer-Use Agent UI
• Get started with the Computer-Use Agent CLI
• Get started with the Python SDKs

Usage (Docs)

1

pip install cua-agent[all]

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14

from agent import ComputerAgent

agent = ComputerAgent(
    model="anthropic/claude-3-5-sonnet-20241022",
    tools=[computer],
    max_trajectory_budget=5.0
)

messages = [{"role": "user", "content": "Take a screenshot and tell me what you see"}]

async for result in agent.run(messages):
    for item in result["output"]:
        if item["type"] == "message":
            print(item["content"][0]["text"])

Output format (OpenAI Agent Responses Format):

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54

{ 
  "output": [
    # user input
    {
        "role": "user",
        "content": "go to trycua on gh"
    },
    # first agent turn adds the model output to the history
    {
        "summary": [
            {
                "text": "Searching Firefox for Trycua GitHub",
                "type": "summary_text"
            }
        ],
        "type": "reasoning"
    },
    {
        "action": {
            "text": "Trycua GitHub",
            "type": "type"
        },
        "call_id": "call_QI6OsYkXxl6Ww1KvyJc4LKKq",
        "status": "completed",
        "type": "computer_call"
    },
    # second agent turn adds the computer output to the history
    {
        "type": "computer_call_output",
        "call_id": "call_QI6OsYkXxl6Ww1KvyJc4LKKq",
        "output": {
            "type": "input_image",
            "image_url": "data:image/png;base64,..."
        }
    },
    # final agent turn adds the agent output text to the history
    {
        "type": "message",
        "role": "assistant",
        "content": [
          {
            "text": "Success! The Trycua GitHub page has been opened.",
            "type": "output_text"
          }
        ]
    }
  ], 
  "usage": {
      "prompt_tokens": 150,
      "completion_tokens": 75,
      "total_tokens": 225,
      "response_cost": 0.01,
  }
}

Computer (Docs)

1

pip install cua-computer[all]

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14

from computer import Computer

async with Computer(
    os_type="linux",
    provider_type="cloud",
    name="your-container-name",
    api_key="your-api-key"
) as computer:
    # Take screenshot
    screenshot = await computer.interface.screenshot()

    # Click and type
    await computer.interface.left_click(100, 100)
    await computer.interface.type("Hello!")

Resources

• How to use the MCP Server with Claude Desktop or other MCP clients - One of the easiest ways to get started with Cua
• How to use OpenAI Computer-Use, Anthropic, OmniParser, or UI-TARS for your Computer-Use Agent
• How to use Lume CLI for managing desktops
• Training Computer-Use Models: Collecting Human Trajectories with Cua (Part 1)

Modules

Community

Join our Discord community to discuss ideas, get assistance, or share your demos!

License

Cua is open-sourced under the MIT License - see the LICENSE file for details.

Portions of this project, specifically components adapted from Kasm Technologies Inc., are also licensed under the MIT License. See libs/kasm/LICENSE for details.

Microsoft’s OmniParser, which is used in this project, is licensed under the Creative Commons Attribution 4.0 International License (CC-BY-4.0). See the OmniParser LICENSE for details.

Third-Party Licenses and Optional Components

Some optional extras for this project depend on third-party packages that are licensed under terms different from the MIT License.

• The optional “omni” extra (installed via pip install "cua-agent[omni]") installs the cua-som module, which includes ultralytics and is licensed under the AGPL-3.0.

When you choose to install and use such optional extras, your use, modification, and distribution of those third-party components are governed by their respective licenses (e.g., AGPL-3.0 for ultralytics).

Contributing

We welcome contributions to Cua! Please refer to our Contributing Guidelines for details.

Trademarks

Apple, macOS, and Apple Silicon are trademarks of Apple Inc.
Ubuntu and Canonical are registered trademarks of Canonical Ltd.
Microsoft is a registered trademark of Microsoft Corporation.

This project is not affiliated with, endorsed by, or sponsored by Apple Inc., Canonical Ltd., Microsoft Corporation, or Kasm Technologies.

Stargazers

Thank you to all our supporters!