Lizard Sandboxes
Cloud Firecracker microVMs for AI agents, with code execution, a browser and desktop, snapshots, and Persistent Volumes.
Tags
Description
Lizard (lizard.build) provides hosted Linux microVMs, each with its own kernel, for running agent code. Agents can run shell commands and stateful Python, read and write files, open private preview URLs, and use a desktop and browser. The Lizard SDK supports TypeScript, JavaScript and Python; a CLI and an HTTP API are also available. Pause and resume preserve disk, memory and running processes; snapshots and fork copy sandbox state. Persistent Volumes mounted at /workspace keep files after a sandbox stops (node-local storage, one sandbox at a time, no fork while attached). Runs in US East and EU West; no GPUs or self-hosting. Paid plans only after a 7-day card-required trial. Per-sandbox egress allow/deny controls are not documented.
Agent access
Related
- E2B: Open-source, secure cloud sandboxes for AI agents: an isolated machine per agent to run code, browse and use tools.
- Daytona: Secure and elastic infrastructure (sandboxes) for running AI-generated code.
- Modal: Cloud platform for production AI: inference, training, sandboxes and serverless functions on one platform.
Guides & comparisons
- Code sandboxes for AI agents (Comparison): E2B, Daytona, Modal, Vercel Sandbox, Cloudflare Sandboxes, Fly.io Sprites and microsandbox compared on isolation, pricing, free tier, lifetime, persistence, egress controls, SDKs, GPUs and self-hosting.
Sources
- https://lizard.build/
- https://lizard.build/pricing
- https://lizard.build/docs/sandboxes/
- https://lizard.build/docs/platform/limits/
- https://lizard.build/llms.txt
- https://github.com/lizard-build/lizard-sdk
- https://github.com/lizard-build/lizard-sdk/blob/main/docs/sandboxes.md
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)