Skip to main content

Welcome to VoidRun

VoidRun is a modern sandbox environment that provisions highly secure, isolated MicroVMs for AI agents and developers. Execute code safely with enterprise-grade security and lightning-fast performance.

πŸš€ Quick Start

Get up and running with VoidRun in under 5 minutes.

πŸ“š Core Concepts

Understand sandboxes, lifecycle, and execution models.

What is VoidRun?

VoidRun provides isolated MicroVM environments designed for AI workloads and developer tooling. Whether you’re building AI agents, code interpreters, or CI/CD pipelines, VoidRun offers the security and flexibility you need.

Key Features

⚑ Performance

Infinite running sandboxes with auto-sleep after 5 minutes of idle time. Resume in under 200ms.

🐳 Flexible Environments

Custom Docker images supported. Run Docker and K3s natively inside sandboxes.

πŸ”’ Security

Granular internet access control with allowlists and blocklists. Native Tailscale VPN connectivity.

πŸ’» Code Execution

Full code interpreter capabilities. Install Ubuntu, Python, and TypeScript packages dynamically.

🌐 Direct Access

Interact with sandboxes via VNC and SSH for debugging and development.

πŸ”‘ Simple Auth

Bearer token authentication via VOIDRUN_API_KEY environment variable or SDK options.

How It Works

VoidRun architecture is built on MicroVMs for maximum isolation and security:
1

Create a Sandbox

Provision an isolated MicroVM with your preferred configuration and base image.
2

Execute Code

Run commands, upload files, and execute code in multiple languages securely.
3

Manage Lifecycle

Sandboxes run infinitely with auto-sleep. Resume instantly when needed.
4

Clean Up

Delete sandboxes when finished, or let them hibernate for later use.

Use Cases

AI Code Execution

Build AI agents that can safely execute generated code in isolated environments:
import { VoidRun } from '@voidrun/sdk';

const vr = new VoidRun({ apiKey: process.env.VOIDRUN_API_KEY });
const sandbox = await vr.createSandbox({ cpu: 2, mem: 2048 });

// Execute AI-generated code safely
const result = await sandbox.runCode(`
def calculate_fibonacci(n):
    if n <= 1:
        return n
    return calculate_fibonacci(n-1) + calculate_fibonacci(n-2)
    
print(calculate_fibonacci(10))
`, { language: 'python' });

console.log(result.stdout);
await sandbox.remove();

Development Environments

Spin up ephemeral development environments with custom tooling:
import { VoidRun } from '@voidrun/sdk';

const vr = new VoidRun({ apiKey: process.env.VOIDRUN_API_KEY });
const sandbox = await vr.createSandbox({ 
  templateId: 'node:20',
  cpu: 4, 
  mem: 4096 
});

// Install dependencies and run tests
await sandbox.exec({ command: 'npm install && npm test' });
await sandbox.remove();

Who is VoidRun For?

AudienceUse Case
AI/ML EngineersBuild code-interpreting AI agents with safe execution
Platform TeamsOffer secure execution environments at scale
DevOps EngineersCreate ephemeral CI/CD environments
Security TeamsRun untrusted code in isolated MicroVMs
SaaS BuildersEmbed code execution in your products

Next Steps

Ready to start building? Check out the Quickstart Guide to create your first sandbox.

SDK Reference

Explore the TypeScript and Python SDKs.

API Reference

View the complete REST API documentation.

Resources

πŸ“Š Dashboard

Manage your sandboxes at dash.void-run.com

πŸ™ GitHub

Star us on GitHub

πŸ’¬ Community

Join our community for support and updates.