Initial pilot: access by invitation, at most 20 sandboxes per organization and
4 vCPU / 8 GiB per sandbox. Read the launch guide for first-run
instructions, cold-image expectations, recovery behavior, support, and pilot terms.
The timings in these docs are historical development measurements, not service guarantees.
Start
Log in athttps://console.nanovm.dev.lithosai.cloud (your first login creates
your organization), create an API key on the API Keys page (shown once,
starts with lith_sk_), then:
pip install nanovm/lithosbox from PyPI — those names are not ours.
Keep one client per process: its connections are reused, which is most of the
difference between a 10 ms create and a 200 ms one. with nv.sandboxes.create() as sb:
deletes on exit (best effort — the downloadable first-run example shows explicit
delete-and-confirm). Complete first run with curl instead:
curl -fsS https://console.nanovm.dev.lithosai.cloud/examples/first-sandbox.sh | bash.
Coding agents: python -m lithosbox skill install, then tell the agent
“set up nanoVM” — details in For coding agents.
Where to go next
Run commands
Shell strings, argv form, sessions, and background work.
Save, restore, fork
Checkpoint a sandbox and fork live copies for parallel work.
The twelve operations
The complete SDK surface — one operation per job, arguments select behavior.
HTTP API
Every route, the sandbox object, and every status code.
Query these docs from your agent
These docs are agent-readable without any setup: append.md to any page URL
for its Markdown, fetch /llms.txt for the index or
/llms-full.txt for every page concatenated, install the
nanoVM skill with npx skills add https://docs.nanovm.dev.lithosai.cloud, or
connect the docs as an MCP server:
Getting help
Report anything odd — a stuck sandbox, a confusing error, a slow create — with thevm_id and the time; every request is traceable on our side. See the
launch guide for support channels and pilot terms.