# Lium GPU Cloud: live GPU prices and rent links

> Live prices and rent links for H100, H200, B200 and RTX 4090 GPUs on Lium, plus Lium docs search.

- name: io.github.Datura-ai/lium-gpu-cloud
- version: 1.0.0
- connection class: R0 (autonomous)
- trust: 97/100
- quality: 78/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.Datura-ai%2Flium-gpu-cloud
- tools: https://api.protogrid.dev/v1/servers/io.github.Datura-ai%2Flium-gpu-cloud/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "lium-gpu-cloud": {
      "type": "http",
      "url": "https://docs.lium.io/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 100
- provenance: 90
- drivers: +repository +namespace-matches-repo +website +reachable +uptime-100% +updated-24d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 78/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 94 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 13 of 13 tools declare readOnlyHint or destructiveHint
- warn `tools.directory_hints`: 2 of 13 tools declare all four hints; missing: list_pods (idempotentHint); get_pod (idempotentHint); search (idempotentHint); …
- pass `tools.token_cost`: about 3,093 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump
- pass `tools.hidden_text`: no hidden text in tool names, descriptions, schemas or annotations

### stability: 50 (weight 15)

- warn `stability.changes`: 13 tool changes in 30 days
- fail `stability.rug_pull`: 2 tools kept their name but changed most of their description in 30 days; review before trusting them
- pass `stability.new_links`: no tool change named a new site

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 13, about 3,093 tokens to load them all
- tool set hash: d7bbd5dae07bcca10e3e71604b10bf76
- tools last changed: 2026-11-13T20:37:57.537Z
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.Datura-ai/lium-gpu-cloud.svg)](https://protogrid.dev/servers/io.github.Datura-ai/lium-gpu-cloud)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-11-13 `remove_pod` changed (description, inputSchema)
  - before: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost. Removes exactly one pod per call, and only with that pod's removal_token from get_pod or list_pods (valid 30 to 60 minutes) and its current pod_name; otherwise nothing is removed.
  - after: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost. Removes exactly one pod per call, and only with that pod's removal_token from prepare_removal (valid 30 to 60 minutes) and its current pod_name; otherwise nothing is removed.
- 2026-11-13 `prepare_removal` added
  - The pods a removal would delete: name, id, GPU, status, price per hour and spend so far, and each pod's removal_token for remove_pod (valid 30 to 60 minutes). Removes nothing. Up to 50 pod ids per call; list_pods gives the ids.
- 2026-11-13 `get_pod` changed (description)
  - before: One pod: status, ssh command (once RUNNING), price per hour, spend so far and its removal_token for remove_pod.
  - after: One pod: status, ssh command (once RUNNING), price per hour and spend so far.
- 2026-11-13 `rent_pod` changed (description, inputSchema)
  - before: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. quoted_price_per_gpu_hour caps the price: a node asking more than one cent per GPU-hour above it is not rented. idempotency_key is required: a call with the same key within 24 hours returns the first rental instead of renting again.
  - after: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. quoted_price_per_gpu_hour caps the price: a node asking more than one cent per GPU-hour above it is not rented. idempotency_key is required: a call with the same key within 24 hours returns the first rental instead of renting again. quote_id is required: the one quote_rent returned for the same gpu or node_id, count, country, template_id and quoted_price_per_gpu_hour; other terms rent nothing.
- 2026-11-13 `quote_rent` added
  - The price and billing terms of a rental before it is made, from the live nodes feed: GPU, count, USD per GPU-hour and per hour, and when billing starts and stops. Rents nothing. Returns quote_id, which rent_pod requires with the same arguments (valid 30 to 60 minutes).
- 2026-11-13 `remove_pod` changed (description, inputSchema) **meaning changed**
  - before: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost. Before calling, read the pod (get_pod or list_pods) and get_balance, then tell the user in your reply, not only in the approval: the pod name and id, its GPU, its price per hour and their current balance, and that data on the pod is lost. Call only after they agree, also when they asked you not to ask; one call per pod.
  - after: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost. Removes exactly one pod per call, and only with that pod's removal_token from get_pod or list_pods (valid 30 to 60 minutes) and its current pod_name; otherwise nothing is removed.
- 2026-11-13 `get_pod` changed (description)
  - before: One pod: status, ssh command (once RUNNING), price per hour and spend so far.
  - after: One pod: status, ssh command (once RUNNING), price per hour, spend so far and its removal_token for remove_pod.
- 2026-11-13 `rent_pod` changed (description, inputSchema, annotations)
  - before: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Before calling, tell the user in your reply, not only in the approval: the GPU model and count (and the node, with node_id), the total price in USD per hour, and that billing continues per second until the pod is removed. Call only after they agree. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. idempotency_key is required: if a call fails or times out, retry with the same key within 24 hours so it cannot rent twice.
  - after: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. quoted_price_per_gpu_hour caps the price: a node asking more than one cent per GPU-hour above it is not rented. idempotency_key is required: a call with the same key within 24 hours returns the first rental instead of renting again.
- 2026-11-09 `remove_pod` changed (description, inputSchema) **meaning changed**
  - before: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost.
  - after: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost. Before calling, read the pod (get_pod or list_pods) and get_balance, then tell the user in your reply, not only in the approval: the pod name and id, its GPU, its price per hour and their current balance, and that data on the pod is lost. Call only after they agree, also when they asked you not to ask; one call per pod.
- 2026-11-09 `rent_pod` changed (description, inputSchema)
  - before: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. idempotency_key is required: if a call fails or times out, retry with the same key within 24 hours so it cannot rent twice.
  - after: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Before calling, tell the user in your reply, not only in the approval: the GPU model and count (and the node, with node_id), the total price in USD per hour, and that billing continues per second until the pod is removed. Call only after they agree. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. idempotency_key is required: if a call fails or times out, retry with the same key within 24 hours so it cannot rent twice.

Full history: https://api.protogrid.dev/v1/servers/io.github.Datura-ai%2Flium-gpu-cloud/changes

## Remotes

- https://docs.lium.io/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (13)

- `estimate_provider_earnings`: Estimate what a provider earns per month on Lium by listing GPUs of a model, as the rewards calculator does: rental income (95% of the renter price) and rented extra at the last 30 days' utilisation, plus idle pay at the validator's rate for every unrented hour; with what those GPUs actually earned and the middle-half range of that potential across providers. Use when someone asks what their GPUs could earn on Lium; for renting GPUs use search_gpus. An estimate, not a guarantee.
- `get_balance`: Your Lium balance in USD.
- `get_gpu_prices`: Live price range (min, median, max USD per GPU-hour) and free GPU count for each GPU model rentable on Lium right now. Use for "what does an H100 cost" or comparing models; use search_gpus to pick a node to rent. Pass count for multi-GPU quotes: a 1-GPU price can come from a node with one free GPU.
- `get_pod`: One pod: status, ssh command (once RUNNING), price per hour and spend so far.
- `list_pods`: Your Lium pods with status, ssh command and price per hour.
- `list_templates`: Docker templates a pod can start from (official ones, your own and other users' public ones; official and yours mark which), for rent_pod template_id.
- `prepare_removal`: The pods a removal would delete: name, id, GPU, status, price per hour and spend so far, and each pod's removal_token for remove_pod (valid 30 to 60 minutes). Removes nothing. Up to 50 pod ids per call; list_pods gives the ids.
- `quote_rent`: The price and billing terms of a rental before it is made, from the live nodes feed: GPU, count, USD per GPU-hour and per hour, and when billing starts and stops. Rents nothing. Returns quote_id, which rent_pod requires with the same arguments (valid 30 to 60 minutes).
- `read_page`: Return the raw Markdown source for a documentation page. Accepts a full URL, absolute slug (e.g. /providers/quickstart), or basename (e.g. quickstart). On tie among basenames, prefers pages in audience order: Providers > Validators > Renters > Developers.
- `remove_pod`: Request pod deletion. Billing stops when accepted; deletion may finish asynchronously (get_pod shows DELETING). Data on the pod is lost. Removes exactly one pod per call, and only with that pod's removal_token from prepare_removal (valid 30 to 60 minutes) and its current pod_name; otherwise nothing is removed.
- `rent_pod`: Rent a GPU pod on Lium with your API key and wait until it runs; returns the pod id and the ssh command. Spends money from your Lium balance: the pod is billed per second until remove_pod, also when idle. Give gpu (and count, optional max_price_per_gpu_hour and country) to let Lium pick the cheapest matching node, or node_id from search_gpus alone for that node. quoted_price_per_gpu_hour caps the price: a node asking more than one cent per GPU-hour above it is not rented. idempotency_key is required: a call with the same key within 24 hours returns the first rental instead of renting again. quote_id is required: the one quote_rent returned for the same gpu or node_id, count, country, template_id and quoted_price_per_gpu_hour; other terms rent nothing.
- `search`: Search the Lium documentation for pages matching the query. Use for how-to questions (renting, SSH, templates, providing GPUs); for live GPU availability or prices use search_gpus or get_gpu_prices instead. Optionally filter by audience (providers, validators, pod-users, developers); without one, pages for the other side (renter or provider, guessed from the query) rank lower.
- `search_gpus`: List GPU nodes that can be rented on Lium right now, cheapest per GPU-hour first, with live USD prices and a rent link for each. Use when the user wants a specific machine (GPU model, count, country, price cap); use get_gpu_prices for a price overview. Each node has tier (secure or spot), reliability_score and min_gpus_per_rental (some nodes rent only whole). Read-only: next_step says how to rent one.