Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.baseten.co/llms.txt

Use this file to discover all available pages before exploring further.

Frontier Gateway delivers a signed webhook to your configured endpoint for each inference request through the gateway. This is a webhook delivered to your endpoint, not an endpoint you call. Delivery is gated to workspaces onboarded to Frontier Gateway. For the consumption guide (verification snippets, retry semantics, and recommended consumption pattern), see Billing webhooks.

​
Event type

Frontier Gateway emits a single event type today: API_BILLING_USAGE. The payload uses the standard Baseten envelope, where type is the discriminator and data holds the event-specific fields.
​
type
string
required
Event type discriminator. Always "API_BILLING_USAGE" for Frontier Gateway billing events.
​
data
object
required
Event-specific payload.
  • events (array, required): One or more event entries per delivery. Each entry corresponds to a single inference request through the gateway. See Event fields below for the entry shape.

​
Event fields

Each entry in data.events[] describes one inference request.
​
idempotencyKey
string
required
Stable identifier for the event. Use this to deduplicate on your side; the same key never describes two distinct events. Stable across retries of the same delivery.
​
timestamp
string
required
ISO 8601 UTC timestamp of the inference request, for example "2025-07-07T23:40:35.905Z".
​
requestId
string
required
Per-request identifier, useful for correlating billing events with platform logs.
​
requestMetadata
object | null
required
Freeform JSON object passed through from the inference request. May be null when no metadata is supplied.
​
modelSlug
string
required
Slug of the model invoked, in your-org/your-model form.
​
externalCustomerId
string
required
The customer_id of the federated user that owns the key used for the request. Set when the user was created via POST /v1/gateway/users.
​
tokens
object
required
Token counts for the request.
  • inputTokens (integer, required): Prompt tokens consumed by the request.
  • outputTokens (integer, required): Generated tokens returned by the model.
  • cachedInputTokens (integer, required): Prompt tokens served from cache, when applicable.

​
Headers

Baseten sets two headers on every delivery.
​
X-Baseten-Signature
string
required
HMAC-SHA256 signature of the raw request body, in v1=<hex> format, computed with your workspace’s webhook signing secret. Verify this on every delivery before trusting the payload. For more information, see Verify the signature.
​
X-Baseten-Request-ID
string
required
UUID generated per outbound delivery. Reused across retry attempts of the same delivery, so it’s useful as a correlation ID against Baseten platform logs but not as a deduplication key. Use idempotencyKey to dedupe events.

​
Signature format

The X-Baseten-Signature header has the format v1=<hex>, where <hex> is the HMAC-SHA256 of the raw request body computed with your workspace’s webhook signing secret. Compute the HMAC against the raw request bytes, not a re-serialized JSON, and compare in constant time. For runnable verification snippets in Python and Node.js, see Verify the signature.

​
Sample payload

{
  "type": "API_BILLING_USAGE",
  "data": {
    "events": [
      {
        "idempotencyKey": "01J9X7Y0Z3K4M5N6P7Q8R9S0T1",
        "timestamp": "2025-07-07T23:40:35.905Z",
        "requestId": "5e4a8c1a-2b3c-4d5e-9f0a-1b2c3d4e5f6a",
        "requestMetadata": {},
        "modelSlug": "your-org/your-model",
        "externalCustomerId": "1",
        "tokens": {
          "inputTokens": 100,
          "outputTokens": 200,
          "cachedInputTokens": 300
        }
      }
    ]
  }
}