Developers

Build on top of Lium programmatically — whether you're a human writing a service or an AI agent driving the platform on a user's behalf. Everything Lium exposes (REST API, CLI, Python SDK, MCP, llms.txt) is documented here. The external URLs in the lists below are stable and safe to bookmark.

For AI agents

Lium is agent-first compute. Start at AI Agents — a single page that wires up all four agent surfaces — then Scripting for agents for the non-interactive contract (--yes, --format json, exit codes, idempotent rent, teardown):

• https://docs.lium.io/mcp — docs MCP server (search + read pages)
• https://docs.lium.io/llms-full.txt — full docs as one markdown bundle
• https://lium.io/api/openapi.json — live OpenAPI 3.1 spec for the platform API
• curl -fsSL https://lium.io/install.sh | bash — shell-driven pod management

For developers

One package, lium.io, ships the CLI and the Python SDK; both read the same key from ~/.lium/config.ini. In order:

Step	What you do	Page
1	Install: pip install lium.io or curl -fsSL https://lium.io/install.sh | bash	CLI installation
2	Authenticate: lium signup (new account) or lium init (existing) — or paste a key from Access → API Keys	lium signup · lium init
3	Rent, connect, run, stop from the shell: lium ls → up → ssh/exec → rm	CLI quickstart · Reference
4	Script the same lifecycle in Python with Lium()	SDK · Pod lifecycle example
5	Run one function on a GPU and get the result back: @lium.machine	SDK → @lium.machine · Remote inference example
6	Call what the CLI does not wrap over REST with X-API-Key	Developer quickstart · API

lium ls and lium ps take --format json; exec, describe, balance, signup, topup create and every lium provider … command take --json, so the CLI is scriptable without the SDK.

• MCP endpoint — Model Context Protocol docs server
• llms.txt — /llms.txt and /llms-full.txt for LLM context

The live Swagger UI is hosted at lium.io/documents; the raw spec lives at lium.io/api/openapi.json.