Sign Up

Daytona Documentation

Start managing your Sandboxes with Daytona.

Daytona is a secure and elastic infrastructure for running AI-generated code.

Daytona provides full composable computers — sandboxes — with complete isolation, a dedicated kernel, filesystem, network stack, and allocated vCPU, RAM, and disk.

Sandboxes are the core component of the Daytona platform, spinning up in under 90ms from code to execution and running any code in Python, TypeScript, and JavaScript. Built on OCI/Docker compatibility, massive parallelization, and unlimited persistence, sandboxes deliver consistent, predictable environments for AI, agent and reinforcement learning workflows.

Agents and developers interact with sandboxes programmatically using the SDKs, API, and CLI.

Agent Prompt
Help me set up Daytona in this project. Before writing any Daytona code, verify the setup: 1. Check that the Daytona SDK is installed for the language used in this project. If not, install it. Python: `pip install daytona` TypeScript: `npm install @daytona/sdk` Ruby: `gem install daytona` Go: `go get github.com/daytona/clients/sdk-go` Java: add the `io.daytona:sdk` dependency to `build.gradle.kts` or `pom.xml` 2. Check DAYTONA_API_KEY in the shell environment or in environment files. If not set, tell the user they need an API key and point them to https://app.daytona.io/dashboard/keys to create one. 3. Write a script that initializes the Daytona client, creates a sandbox, runs a "Hello World" program inside it, captures the output, and deletes the sandbox. 4. Run the script to verify the sandbox works.

Get started

  1. Create an account → app.daytona.io
  2. Get an API key → app.daytona.io/dashboard/keys
  3. Save the API key → DAYTONA_API_KEY environment variable or in a .env file.
DAYTONA_API_KEY=YOUR_API_KEY

  1. Install the Python SDK

    pip install daytona

  2. Create a sandbox and run code

    # Import the Daytona SDK
from daytona import Daytona, DaytonaConfig

# Define the configuration
config = DaytonaConfig(api_key="YOUR_API_KEY") # Replace with your API key

# Initialize the Daytona client
daytona = Daytona(config)

# Create the Sandbox instance
sandbox = daytona.create()

# Run code
response = sandbox.process.code_run('print("Hello World")')
print(response.result)

Next steps

Daytona provides core platform resources for managing sandboxes, environments, and storage.

Linux sandboxes
Linux container and virtual machine runtime for running processes, executing code, and Linux-specific tools.
Windows sandboxes
Windows OS runtime in a virtual machine for running Windows applications and tooling.
macOS sandboxes
macOS runtime for automating applications and desktop interactions.
GPU sandboxes
NVIDIA and AMD GPU runtime for CUDA development, 3D and graphics workloads, and other GPU-accelerated compute.

Reference

Daytona provides a comprehensive set of resources for developers and agents to get started and build on the platform. You can find the following resources in the documentation:

On this page

Get startedNext stepsReference

Computersfor Agents

Isolated sandboxes for agent code,
ready in under 100 ms.

Start for Free