Network stats

Public read-only snapshot of the live pool.

GET https://api.malibu.tech/v1/stats/overview

No API key. Cached about 30 seconds (stale_after in the body). Same payload on https://malibu.tech/v1/stats/overview. GET /v1/network-stats is a compatibility alias for the same overview body. Prefer /v1/stats/overview.

Response (abridged)

{
  "generated_at": "2026-08-16T06:45:13Z",
  "stale_after": "2026-08-16T06:45:43Z",
  "network": {
    "tokens_served_total": 7332222,
    "tokens_in_total": 2105292,
    "tokens_out_total": 5226930,
    "requests_total": 28995,
    "nodes_online": 4,
    "nodes_hardware_attested": 0,
    "models_serving": 2,
    "gpu_cores_total": 50,
    "cpu_cores_total": 40,
    "unified_ram_gb_total": 80,
    "avg_tokens_per_request": 252
  },
  "timeseries": {
    "rpm_30m": {
      "bucket_seconds": 60,
      "points": [{ "t": "2026-08-16T06:15:00Z", "value": 3 }]
    }
  }
}

Example values are a dated snapshot (16 Aug 2026). Live numbers change; call the endpoint.

Fields

Field	Meaning
`generated_at`	RFC3339 snapshot time
`stale_after`	Treat the payload as expired after this timestamp
`network.nodes_online`	Providers with an active connection at snapshot time
`network.models_serving`	Distinct models with at least one warm provider
`network.requests_total`	Lifetime billed-request count on this surface
`network.tokens_served_total`	Lifetime prompt + completion tokens
`network.tokens_in_total` / `tokens_out_total`	Lifetime prompt vs completion split
`network.nodes_hardware_attested`	Providers with hardware attestation — 0 today (not live)
`network.unified_ram_gb_total`	Sum of reported unified memory
`timeseries.rpm_30m`	Requests per minute, 60 s buckets

Capacity fields (bandwidth_gb_per_s, network_power_kw, core counts) are estimates from hardware profiles. The payload includes capacity_estimate_sources naming the method. Do not treat them as metered telemetry. Network-wide decode tokens/sec is not published on this endpoint. The console bill line can show client-measured tok/s for a single stream.

Rate card (separate endpoint)

GET https://api.malibu.tech/v1/rate-card

Same payload on https://malibu.tech/v1/rate-card. See Pricing comparison and Model catalog.

Warm models (authenticated)

GET https://api.malibu.tech/v1/models
Authorization: Bearer <MALIBU_API_KEY>

Returns only models with at least one warm provider. See Models.

Operations → Network stats — how aggregates are computed
Network status — live vs planned ledger
Reputation — how per-provider signals surface

Litepaper

Getting started

Buyers

Providers

Network & trust

Guides index

Ledger

Network stats

Network stats

Response (abridged)

Fields

Rate card (separate endpoint)

Warm models (authenticated)

​Network stats

​Response (abridged)

​Fields

​Rate card (separate endpoint)

​Warm models (authenticated)

​Related

Network stats

Response (abridged)

Fields

Rate card (separate endpoint)

Warm models (authenticated)

Related