Daytona

Open-source, secure, and elastic infrastructure for running AI-generated code, offering sub-90ms sandbox provisioning, LSP support, and persistent environments.

Daytona is an execution infrastructure designed specifically for AI-generated code, addressing security isolation and persistence challenges for AI agents.

Core Capabilities#

Infrastructure Layer:

Sub-90ms sandbox provisioning, from code to execution
Fully isolated runtime environment, zero-risk AI code execution
Support for large-scale concurrent AI workflows
Indefinitely persistent sandboxes for long-running tasks and stateful agents
OCI/Docker compatible, supports any image for environment creation

Programmatic Control:

FileSystem: Full CRUD operations with granular permission control
Native Git: Built-in Git operations and secure credential handling
LSP Support: Built-in language server for multi-language completion and real-time analysis
Process Execution: Execute code and commands in isolated environments with real-time output streaming

Advanced Features:

Environment Snapshots: Instantly save and restore agent workflows
Multi-region Deployment: US East/West, EU Central/West, India South
Volume Management: Share data across sandboxes without breaking isolation
Computer Use: Programmatic control of Linux/macOS/Windows desktop environments

Human Collaboration Tools:

Instant secure SSH access
VS Code Browser one-click open
Web Terminal: Full terminal in browser

Installation & Quick Start#

Python SDK: pip install daytona

TypeScript SDK: npm install @daytonaio/sdk

Go SDK: import "github.com/daytonaio/daytona/libs/sdk-go/pkg/daytona"

Minimal Example:

from daytona import Daytona, DaytonaConfig
config = DaytonaConfig(api_key="YOUR_API_KEY")
daytona = Daytona(config)
sandbox = daytona.create()
response = sandbox.process.code_run('print("Hello World")')
print(response.result)
sandbox.delete()