Skip to main content
GET
Get deployment metrics timeseries

Authorizations

Authorization
string
header
required

API key with the vk_ prefix. Create one from the dashboard.

Path Parameters

deployment_id
string
required

Deployment ID

Query Parameters

from
string<date-time>
to
string<date-time>
interval
enum<string>
Available options:
auto,
minute,
five_minutes,
hour,
day
replica_id
string

Scope buckets, summary, and cost to one replica of the deployment (404 when the replica doesn't belong to it). Omitted = deployment-wide.

Response

Server-filled inference metrics buckets

object
string
required
deployment_id
string
required
from
string<date-time>
required
to
string<date-time>
required
interval
enum<string>
required
Available options:
auto,
minute,
five_minutes,
hour,
day
summary
object
required
buckets
object[]
required
replica_id
string | null

Set when the response is scoped to one replica (?replica_id=); absent for deployment-wide scope.

cost_efficiency
null | object