hosts

Market stats (host): per-GPU-model supply, listed-rate band, fill rate and time-to-first-rental — aggregate only

Our own marketplace, never a competitor's. Aggregate only: no machine id, host id, offer id or any individual host's rate is ever returned — the only rate values that escape are the p10/median/p90 band of a bucket of at least min_sample_size listed machines, plus the CALLER's own median listed rate (your_rate_micro). Below the threshold a bucket reports its sample_size with every metric null; a median is never computed from a handful of observations. median_time_to_first_rental_s has its own sample gate (rental_sample_size), because a model can be widely listed and barely rented. Window is half-open UTC [from, to).

get/api/v1/hosts/market

Request

  • Base URL: https://amplerun-api.spiry.org
  • URL: https://amplerun-api.spiry.org/api/v1/hosts/market
  • Auth: HTTP bearer

Query parameters

gpu_modelstring

Exact manifest GPU model to restrict to.

windowstring

Trailing window, e.g. 30d (1..365); anything else is 422.

Response

Per-model market aggregates.

window_daysinteger required
fromstring date-time required
tostring date-time required
min_sample_sizeinteger required

Observations required before any metric is reported.

Example response

{
  "from": "2026-09-25T12:00:00Z",
  "to": "2026-09-25T12:00:00Z",
  "models": [
    {
      "rate_micro": {
        "p10_micro": "1",
        "median_micro": "1",
        "p90_micro": "1"
      },
      "median_time_to_first_rental_s": "1",
      "your_rate_micro": "1"
    }
  ]
}

Changes

No recorded changes to this operation across all 1 revision of this API.