Quick Start
AerolVM is open-source infrastructure for running isolated code environments. Each sandbox is a fully isolated compute unit with its own filesystem, network stack, and allocated vCPU, RAM, and disk with a support of Docker, GVisor, Kata and Wasm.
The idea is to provide a seamless, secure sandbox experience that can start in under 50ms to 1s.
Sandboxes are the core component of the Aerol.ai platform, spinning up in under 70ms 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 agent workflows.
Agents and developers interact with sandboxes programmatically using the AerolVM SDKs, API, and CLI. Operations span sandbox lifecycle management, filesystem operations, process and code execution, and runtime configuration. Our stateful environment snapshots enable persistent agent operations across sessions, making AerolVM the ideal foundation for AI agent architectures.
Runtimes
Section titled “Runtimes”AerolVM supports multiple container runtimes, giving you a choice between security and compatibility:
| Runtime | Status | Use Case |
|---|---|---|
| Docker | ✅ | Fast startup, broad image compatibility, standard workloads |
| GVisor | ✅ | Kernel-level isolation without a full VM - ideal for untrusted code |
| Firecracker | ✅ | Kernel-level isolation Full VM isolation with hardware virtualization |
| WebAssembly | ✅ | Ultra-lightweight WASI workloads with optional durable checkpoints |
Docker is the default runtime today. GVisor, Firecracker, and WebAssembly are also available when enabled in server config.
Use Cases
Section titled “Use Cases”- AI code execution - run LLM-generated code safely in isolated environments
- CI / ephemeral build agents - spin up a fresh environment per job, destroy when done
- Interactive developer environments - persistent workspaces with SSH, port previews, and file sync
- Data processing pipelines - attach cloud storage, run transforms, and extract results
Next Steps
Section titled “Next Steps”- Server Setup - self-host AerolVM on your own infrastructure
- Create Sandbox - lifecycle states, parameters, and configuration options
- Environment - image selection, env vars, and resource limits
- Streaming Exec - stream stdout/stderr live and use interactive PTY sessions
- Sessions - persistent terminal sessions that survive reconnects
- WebAssembly Sandboxes - create and run WASI modules with durable checkpoints