Skip to main content

Overview

Welcome to the Lium.io Provider Portal documentation. The Provider Portal gives GPU providers a UI to register, monitor, and manage their nodes.

Key features​

  • Flexible management: full control over your machines and availability
  • Pause new rentals: stop a rented node from accepting future rentals without interrupting the current renter
  • Maintenance scheduling: plan downtime without penalties
  • Network-wide visibility: see every node on the platform alongside your own
  • Grafana integration: detailed performance dashboards on each node's detail page

Getting started​

Before you begin, ensure you have:

  1. A provider account — either an e-mail / Google account (nothing to register; create it at provider.lium.io/signup) or your own hotkey registered on subnet 51 — see Authentication
  2. Provider running (hotkey accounts only) — either a self-hosted provider or the Lium.io Central Provider Server opt-in; an e-mail / Google account is connected automatically
  3. Wallet extension (hotkey accounts only) — any SS58-compatible browser extension (e.g. Polkadot.js extension or Bittensor Wallet extension) with your registered provider hotkey imported, used to sign in to the portal
  4. Nodes — at least one GPU node running. See Node Quickstart.

The Provider Portal is organized into several sections:

  • Overview (e-mail / Google accounts) — this month's and all-time earnings from rentals and idle pay, your nodes, and the install command for a new GPU host
  • Overview (hotkey accounts) — the portal's front page (/) once you are signed in with your own key. Three levels are kept apart. Account: your penalties of the last 14 days as a share of your rental income, against the 20% line at which every node of yours moves to Spot (the same numbers the platform decides on). Node: the nodes that are losing you money, each with what is wrong, what to do and about what a day of it costs you — for its GPUs that are idle or not billed; a node whose every GPU is rented and billing shows what it earned. A rented GPU counts as billed only while its rental is charged: the rental is running and the node is active. A rental that is still starting, has failed, or sits on an inactive node counts as a cost. Pay while free: whether each node with a free GPU (a partly rented one included) is paid for staying available, and if not, why. Then what you earned in the last 30 days (rentals after the platform fee, the incentive for staying available, penalties, day by day), your rental payouts (last, next, today so far, each with its time in UTC and a link to the day), and one table of your nodes with status, load, uptime, pay while free, your price against the cap and the average, and what each earned; the table's toolbar carries the same two sync buttons and the Cluster status switch as the Nodes page, and a column sorts ascending on the first click, descending on the second. On a phone every block is one column and the table scrolls inside its card. With no nodes yet, the page shows what machines of each type earn today, from the Rewards Calculator.
  • Nodes — manage your registered GPU machines (status, pricing, notice periods, notifications) and view all nodes across the platform
  • Add Node — register a new GPU node
  • Earnings (hotkey accounts) — rental fees and subnet incentive together, by day and by node, with on-chain settlement proofs
  • Provider Payouts (hotkey accounts) — earnings from rental fees, broken down by day, validator, and node
  • Machine Requests — pending rental inquiries from renters that you can match to your nodes
  • Rewards Calculator (hotkey accounts) — what GPUs of each type earned on Lium over the last 30 days, scaled to the count you enter; see Rewards calculator

Next steps​

  1. Sign in — with e-mail / Google, or with your provider hotkey
  2. Add your first node
  3. Connect Discord so renters can request private pod support channels
  4. Monitor performance
  5. Track payouts