Lium Public API

The Public API from Lium — 1 operation(s) for public.

GitHub OpenAPI

Operations 1

GET /public/provider-economics What GPUs of each class earn on Lium, from the last 30 days #

Work with this as data

Every API here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for apis

7 MCP tools reach this
  • find_apisBrowse and filter every API in the catalog.
  • get_api_artifactsOne API's artifacts, grouped by type.
  • get_openapiThe primary OpenAPI for this API.
  • find_similar_apisAPIs that look like this one.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This API
curl "https://apis.io/api/v1/apis/lium-io:lium-io-public-api"
All apis
curl "https://apis.io/api/v1/apis?limit=25"

Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.

Get an API key

Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. By signing up you are subscribing to the APIs.io and API Evangelist newsletters — you can unsubscribe from any issue. See our Privacy Policy and Terms.

A second provider on the same verified email joins the account you already have.

OpenAPI Specification

lium-io-public-api-openapi.yml Raw ↑
openapi: 3.2.0
info:
  title: Lium Miner Portal Public API
  version: 0.1.0
servers:
- url: /api
tags:
- name: Public
paths:
  /public/provider-economics:
    get:
      tags:
      - Public
      summary: What GPUs of each class earn on Lium, from the last 30 days
      description: 'Per GPU class, measured over the last 30 days: the hour-weighted rental price, rented and idle GPU-hours, utilisation, idle pay and rented extra from the validator ledger, and the live node snapshot. `range` is not measured pay: it is what providers can earn with full idle pay and rented extra on every listed hour (potential, not actual payouts), as interpolated percentiles. No authentication; recomputed at most once an hour.'
      operationId: get_provider_economics_public_provider_economics_get
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ProviderEconomicsResponse'
components:
  schemas:
    ProviderEconomicsRange:
      properties:
        p10_usd_per_gpu_month:
          anyOf:
          - type: number
          - type: 'null'
          title: P10 Usd Per Gpu Month
          description: Null below ten providers
        p25_usd_per_gpu_month:
          type: number
          title: P25 Usd Per Gpu Month
        median_usd_per_gpu_month:
          anyOf:
          - type: number
          - type: 'null'
          title: Median Usd Per Gpu Month
          description: Null below ten providers
        p75_usd_per_gpu_month:
          type: number
          title: P75 Usd Per Gpu Month
        p95_usd_per_gpu_month:
          anyOf:
          - type: number
          - type: 'null'
          title: P95 Usd Per Gpu Month
          description: The top of the class; null below ten providers
        providers:
          type: integer
          title: Providers
          description: Providers with at least a week of one GPU listed that the range is computed over
      type: object
      required:
      - p25_usd_per_gpu_month
      - p75_usd_per_gpu_month
      - providers
      title: ProviderEconomicsRange
      description: 'The spread of the per-GPU-month potential across providers of the class:

        p25–p75, and with ten or more providers p10, median and p95 for the box plot.

        Potential, not payout: each provider''s figure counts every unrented hour at the

        validator''s idle rate and every rented hour at the class''s rented extra (the

        calculator''s basis), never less than the provider was paid. Every value is an

        interpolated percentile (a whole position blends two neighbours), never one

        provider''s own figure, rounded to two significant figures (7,172 -> 7,200) so no

        combination of them solves for one, tied providers included. There is one top

        percentile (p95), no p90.'
    ProviderEconomicsResponse:
      properties:
        as_of:
          type: string
          title: As Of
          description: UTC ISO-8601 time the numbers were computed
        window_days:
          type: integer
          title: Window Days
        hours_per_month:
          type: integer
          title: Hours Per Month
        provider_rental_share:
          type: number
          title: Provider Rental Share
          description: Share of each rental dollar paid to the provider
        classes:
          items:
            $ref: '#/components/schemas/ProviderEconomicsClass'
          type: array
          title: Classes
        unmeasured:
          items:
            $ref: '#/components/schemas/ProviderEconomicsClass'
          type: array
          title: Unmeasured
          description: 'supported models with no published class (no or too little 30-day history): zero measured figures, the validator''s idle rate when it pays one; the rewards calculator estimates them, the measured tables leave them out'
      type: object
      required:
      - as_of
      - window_days
      - hours_per_month
      - provider_rental_share
      - classes
      title: ProviderEconomicsResponse
    ProviderEconomicsClass:
      properties:
        gpu_model:
          type: string
          title: Gpu Model
          description: GPU name as the validator reports it
        rental_price_usd_per_gpu_hour:
          anyOf:
          - type: number
          - type: 'null'
          title: Rental Price Usd Per Gpu Hour
          description: 30-day hour-weighted average rental price per GPU-hour; null when the class had no completed rental in the window
        rented_gpu_hours_30d:
          type: number
          title: Rented Gpu Hours 30D
        idle_gpu_hours_30d:
          type: number
          title: Idle Gpu Hours 30D
        utilisation:
          type: number
          title: Utilisation
          description: rented GPU-hours / (rented + idle GPU-hours), 0–1
        idle_pay_usd_30d:
          type: number
          title: Idle Pay Usd 30D
        rented_extra_usd_30d:
          type: number
          title: Rented Extra Usd 30D
        idle_pay_usd_per_idle_gpu_hour:
          anyOf:
          - type: number
          - type: 'null'
          title: Idle Pay Usd Per Idle Gpu Hour
          description: idle pay ÷ every listed idle GPU-hour, paid or not; the rate the month figure is built on; null when no GPU of the class sat idle
        rented_extra_usd_per_rented_gpu_hour:
          anyOf:
          - type: number
          - type: 'null'
          title: Rented Extra Usd Per Rented Gpu Hour
        idle_rate_usd_per_gpu_hour_max:
          anyOf:
          - type: number
          - type: 'null'
          title: Idle Rate Usd Per Gpu Hour Max
          description: 'the validator''s idle rate for the class, USD per GPU-hour (lium-core machine_prices at the pinned version, RTX PRO 6000 Server Edition rated as the Workstation Edition): what an unrented GPU earns per hour while the node qualifies; the calculator applies it to every unrented hour; null when the validator pays no idle incentive for the class'
        rental_share_usd_per_gpu_month:
          type: number
          title: Rental Share Usd Per Gpu Month
          description: provider_rental_share × rental price × hours_per_month × utilisation
        idle_pay_usd_per_gpu_month:
          type: number
          title: Idle Pay Usd Per Gpu Month
          description: idle rate × hours_per_month × (1 − utilisation)
        rented_extra_usd_per_gpu_month:
          type: number
          title: Rented Extra Usd Per Gpu Month
          description: rented-extra rate × hours_per_month × utilisation
        total_usd_per_gpu_month:
          type: number
          title: Total Usd Per Gpu Month
        live_nodes:
          type: integer
          title: Live Nodes
        live_gpus:
          type: integer
          title: Live Gpus
        live_rented_gpus:
          type: integer
          title: Live Rented Gpus
        range:
          anyOf:
          - $ref: '#/components/schemas/ProviderEconomicsRange'
          - type: 'null'
          description: potential per GPU-month at full incentives, not payout; null when fewer providers than the privacy floor listed the class in the window
        paid_with:
          anyOf:
          - type: string
          - type: 'null'
          title: Paid With
          description: 'unmeasured rows only: a published class in the same scorer family with the same idle-rate ceiling (e.g. the RTX PRO 6000 Max-Q and the Server Edition); a comparison, not a promise that this model earns the same measured figures'
      type: object
      required:
      - gpu_model
      - rental_price_usd_per_gpu_hour
      - rented_gpu_hours_30d
      - idle_gpu_hours_30d
      - utilisation
      - idle_pay_usd_30d
      - rented_extra_usd_30d
      - idle_pay_usd_per_idle_gpu_hour
      - rented_extra_usd_per_rented_gpu_hour
      - idle_rate_usd_per_gpu_hour_max
      - rental_share_usd_per_gpu_month
      - idle_pay_usd_per_gpu_month
      - rented_extra_usd_per_gpu_month
      - total_usd_per_gpu_month
      - live_nodes
      - live_gpus
      - live_rented_gpus
      - range
      title: ProviderEconomicsClass
  securitySchemes:
    JwtAccessBearer:
      type: http
      scheme: bearer
    HTTPBearer:
      type: http
      scheme: bearer