Lium Public nodes feed API
The Public nodes feed API from Lium — 1 operation(s) for public nodes feed.
The Public nodes feed API from Lium — 1 operation(s) for public nodes feed.
Every API here is available over the APIs.io API and to AI agents over MCP.
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
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.curl "https://apis.io/api/v1/apis/lium-io:lium-io-public-nodes-feed-api"
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.
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: 3.2.0
info:
title: Lium Backend Public nodes feed API
description: '## 🚀 Lium Backend API
A comprehensive compute platform that enables users to rent GPU resources from a network of providers/nodes.
*Built with FastAPI, PostgreSQL and Redis.*'
contact:
name: Lium
url: https://lium.io/
email: support@lium.io
license:
name: MIT
url: https://opensource.org/licenses/MIT
version: 1.0.0
servers:
- url: /api
description: Production API
tags:
- name: Public nodes feed
paths:
/public/v1/nodes:
get:
tags:
- Public nodes feed
summary: Public GPU inventory and pricing feed
description: Every node currently rentable on Lium, with normalized GPU names, per-GPU and per-node hourly prices, location and a deep link to rent it. No authentication required. With include_rented=true it also lists the nodes that are on the market but rented now, with available_now false and available_gpu_count 0, so every ask is visible. Fields are only ever added to this response; a breaking change ships as /public/v2, and a retired version stays served for six months after the deprecation notice.
operationId: list_nodes_public_v1_nodes_get
parameters:
- name: include_rented
in: query
required: false
schema:
type: boolean
description: Also list nodes that are on the market but rented now (available_now false).
default: false
title: Include Rented
description: Also list nodes that are on the market but rented now (available_now false).
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/NodesResponse'
'422':
description: Validation Error
content:
application/json:
schema:
$ref: '#/components/schemas/HTTPValidationError'
components:
schemas:
ProviderRating:
properties:
average:
type: number
title: Average
description: Average 1-5 star rating, one decimal.
count:
type: integer
title: Count
description: How many ratings the average is over.
type: object
required:
- average
- count
title: ProviderRating
NodesResponse:
properties:
generated_at:
type: string
format: date-time
title: Generated At
nodes:
items:
$ref: '#/components/schemas/Node'
type: array
title: Nodes
count:
type: integer
title: Count
readOnly: true
type: object
required:
- generated_at
- nodes
- count
title: NodesResponse
description: The feed envelope. `generated_at` is what lets a consumer detect that we went stale.
HTTPValidationError:
properties:
detail:
items:
$ref: '#/components/schemas/ValidationError'
type: array
title: Detail
type: object
title: HTTPValidationError
Node:
properties:
id:
type: string
format: uuid
title: Id
gpu_model:
anyOf:
- type: string
- type: 'null'
title: Gpu Model
gpu_count:
anyOf:
- type: integer
- type: 'null'
title: Gpu Count
available_gpu_count:
anyOf:
- type: integer
- type: 'null'
title: Available Gpu Count
available_now:
type: boolean
title: Available Now
description: True when the node can be rented now. Rows added by include_rented=true are false.
default: true
price_per_gpu_hour:
anyOf:
- type: number
- type: 'null'
title: Price Per Gpu Hour
min_rentable_gpu_count:
anyOf:
- type: integer
- type: 'null'
title: Min Rentable Gpu Count
gpu_memory_gb:
anyOf:
- type: integer
- type: 'null'
title: Gpu Memory Gb
cpu_count:
anyOf:
- type: integer
- type: 'null'
title: Cpu Count
description: CPUs a whole-host rental gets (the host keeps a reserve core); a rental of fewer than gpu_count GPUs gets a proportional share.
ram_gb:
anyOf:
- type: integer
- type: 'null'
title: Ram Gb
description: Whole-host total; a rental of fewer than gpu_count GPUs gets a proportional share.
disk_gb:
anyOf:
- type: integer
- type: 'null'
title: Disk Gb
description: Whole-host total; a rental of fewer than gpu_count GPUs gets a proportional share.
cpu_model:
anyOf:
- type: string
- type: 'null'
title: Cpu Model
description: CPU model string as the host reports it, at most 128 characters.
network_upload_mbps:
anyOf:
- type: number
- type: 'null'
title: Network Upload Mbps
description: Measured upload speed in Mbps, 1 decimal; null until measured.
network_download_mbps:
anyOf:
- type: number
- type: 'null'
title: Network Download Mbps
description: Measured download speed in Mbps, 1 decimal; null until measured.
uptime_minutes:
anyOf:
- type: integer
- type: 'null'
title: Uptime Minutes
description: Minutes our validator has seen this node online, 15 per validation cycle since it was first listed. Cumulative online time, not the current streak.
reliability_score:
anyOf:
- type: number
- type: 'null'
title: Reliability Score
description: Provider reliability over a rolling 14-day window of rentals other than voluntary spot, 0-100, one decimal, recomputed every 12 hours; the same for every node of one provider. Counts penalized rentals and rentals that failed to start because of the provider's node. A failed start does not count when it failed on the renter's own image (building, pulling or running it, or the registry login); when the renter deleted the pod before it started; when it failed on Lium's side before the request reached the node; or when it was a reboot of a pod that had started. null = neither a billed rental nor a counted failed start in the window, so no score.
provider_spot_avg_uptime_hours:
anyOf:
- type: number
- type: 'null'
title: Provider Spot Avg Uptime Hours
description: 'Spot nodes only: average hours this provider''s Spot nodes stayed up before going down, over the last 7 days. null = not Spot, or under 24 h of Spot history.'
provider_rating:
anyOf:
- $ref: '#/components/schemas/ProviderRating'
- type: 'null'
description: The average rating renters gave this provider's nodes when removing pods in the last 90 days; the same for every node of one provider. null = under 5 ratings.
provider_notes:
items:
type: string
type: array
title: Provider Notes
description: Notes the provider shows renters on this node, each from a fixed list (for example "May reboot occasionally"). Empty = none.
disk_type:
anyOf:
- type: string
- type: 'null'
title: Disk Type
description: nvme, ssd, hdd, or unknown when our validator looked and could not tell (a virtual or stacked device). null until our validator has reported it.
tier:
anyOf:
- type: string
- type: 'null'
title: Tier
country:
anyOf:
- type: string
- type: 'null'
title: Country
country_code:
anyOf:
- type: string
- type: 'null'
title: Country Code
region:
anyOf:
- type: string
- type: 'null'
title: Region
city:
anyOf:
- type: string
- type: 'null'
title: City
rent_url:
type: string
title: Rent Url
disk_free_gb:
anyOf:
- type: integer
- type: 'null'
title: Disk Free Gb
description: Free disk on the whole host at the last scrape, GB; null until reported.
gpu_driver:
anyOf:
- type: string
- type: 'null'
title: Gpu Driver
description: NVIDIA driver build as the host reports it, e.g. 570.133.20; at most 32 characters.
cuda_driver:
anyOf:
- type: string
- type: 'null'
title: Cuda Driver
description: Highest CUDA version the installed driver supports, major.minor, e.g. 12.8; null until reported.
gpu_interconnect:
anyOf:
- type: string
- type: 'null'
title: Gpu Interconnect
description: nvlink when every GPU pair on the node is joined by NVLink, confirmed by the validator's P2P and link-state readings; pcie when at least one pair is not on NVLink or a reading contradicts it; null on a single-GPU node, until the topology has been scraped, or when a confirming reading is missing.
region_name:
anyOf:
- type: string
- type: 'null'
title: Region Name
description: Name of the subdivision `region` is the code of.
last_verified_at:
anyOf:
- type: string
format: date-time
- type: 'null'
title: Last Verified At
description: When our validator last passed this node, UTC. It re-checks a listed node about every 15 minutes; null when no passed check is on record.
timezone:
anyOf:
- type: string
- type: 'null'
title: Timezone
description: IANA time zone of the node's city, e.g. Asia/Tokyo.
price_per_node_hour:
anyOf:
- type: number
- type: 'null'
title: Price Per Node Hour
readOnly: true
type: object
required:
- id
- rent_url
- price_per_node_hour
title: Node
description: One rentable node as an aggregator sees it. No fleet internals belong in this model.
ValidationError:
properties:
loc:
items:
anyOf:
- type: string
- type: integer
type: array
title: Location
msg:
type: string
title: Message
type:
type: string
title: Error Type
type: object
required:
- loc
- msg
- type
title: ValidationError
securitySchemes:
JwtAccessBearer:
type: http
scheme: bearer
APIKeyHeader:
type: apiKey
in: header
name: X-API-Key