Developers

Developer API

Authenticate with an API key generated from your wallet account. Usage is charged to your compute account and metered as external revenue.

API keys

Reference

Endpoint

POST /api/v1/inference

Authorization: Bearer lz_live_…
model: registry id (see /models)
prompt: string, ≤ 32k chars
system: optional string
temperature: 0–2, default 0.7
maxTokens: 1–4096, default 512
idempotencyKey: optional, 8–120 chars
Rate limit: 60/min per key, 30/min per IP → 429

Request

curl -X POST https://<your-host>/api/v1/inference \
  -H "Authorization: Bearer lz_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.1-8b-instruct",
    "prompt": "Explain community-owned compute in one paragraph.",
    "temperature": 0.7,
    "maxTokens": 256
  }'

Response

{
  "id": "cm…",
  "model": "llama-3.1-8b-instruct",
  "output": "…",
  "usage": { "input_tokens": 21, "output_tokens": 180, "credits_used": "20100000" },
  "status": "completed",
  "latency_ms": 842,
  "created_at": "2026-09-13T10:00:00.000Z"
}

Errors

401 UNAUTHORIZED         invalid or revoked key
402 INSUFFICIENT_CREDITS  reserve more than available
409 MODEL_UNAVAILABLE     disabled or in maintenance
424 PROVIDER_NOT_CONFIGURED
429 RATE_LIMITED          Retry-After header set
502 PROVIDER_ERROR        upstream failure, reservation released