Hopx

Secure Linux Micro-VMs for AI

Multi-agent systems DEVELOPER TOOLS pay-as-you-go serverless autonomous agents Code interpreter desktop automation Python SDK Jupyter Notebooks Secure Code Execution Cloud compute AI agents sandbox Untrusted code execution Isolated sandbox Micro-virtual machine Firecracker AI code execution Secure sandbox environment Execute arbitrary code Sandbox API Code execution API AI infrastructure agents Linux micro-VMs JavaScript SDK Go SDK Persistent state Real-time streaming Long-running jobs

More

Hopx Product Information

What is Hopx?

Hopx delivers fully isolated Linux micro-VMs that launch in approximately 100 milliseconds, making them perfect for safely scaling AI agents, notebooks, or untrusted code. Each sandbox leverages Firecracker microVMs to ensure hardware-level security and kernel isolation, while offering full state persistence and SDK support for Python, JavaScript, Go, .NET, Java, and PHP. It eliminates cold-start delays and imposes no runtime restrictions, guaranteeing optimal performance and security for AI workloads and dynamic applications.

How to use Hopx?

Developers can integrate Hopx using its SDKs for Python, JavaScript/TypeScript, Go, .NET, Java, and PHP. The workflow includes installing the appropriate SDK (for example, via `npm install @hopx-ai/sdk`), creating a sandbox instance (such as `Sandbox.create({ template: 'code-interpreter' })`), running code inside the sandbox using methods like `runCode()`, and then shutting down the sandbox. The SDK facilitates various actions, including code execution, live output streaming, file management, shell commands, process control, and desktop automation.

Hopx's Core Features

Lightning-fast sandboxes with ~100ms startup

Hardware-level security and kernel isolation using Firecracker microVMs

Unlimited runtime with complete state persistence

Supported SDKs for Python, JavaScript, Go, .NET, Java, and PHP

Code execution, output streaming, file handling, commands, and process management

Desktop automation and live performance monitoring

Hopx's Use Cases

Safely run untrusted code in isolated microVMs

Launch and orchestrate AI agents that write and execute code

Spin up Jupyter notebooks for data analysis

Develop and operate deep research agents continuously

Coordinate autonomous AI workflows with code execution and tools

Manage full cloud desktops via computer use agents

Run background automations, workers, and schedulers

Train and evaluate reinforcement learning agents

Host secure MCP servers in isolated environments

Execute long-running jobs without arbitrary runtime limits

Coordinate multi-agent networks across isolated micro-VMs

FAQ from Hopx

What is a Bunnyshell Sandbox?

A Bunnyshell Sandbox (Hopx) is a fully isolated Linux micro-VM that launches in roughly 100 milliseconds, ideal for securely operating AI agents, notebooks, or untrusted code at scale.

How is it different from containers or serverless functions?

Hopx utilizes Firecracker microVMs, delivering hardware-level security and kernel isolation that far exceeds the safety of containers or serverless functions.

Do I need a credit card to get started?

No credit card is needed to begin with Hopx. You can start for free and receive $200 in complimentary credits.

Hopx Support Email & Customer service contact & Refund contact etc.

For additional contact options, visit the contact us page(http://bunnyshell.com/contact)

Hopx Screenshot