Skip to main content
PUT
Replace API key inference limits

Authorizations

Authorization
string
header
required

Use a JWT bearer token or a Veryfront API key in the Authorization header.

Path Parameters

id
string<uuid>
required

Body

application/json
allowed_models
string[] | null

Allowed catalog model IDs from GET /ai/models. Aliases are normalized to <vendor>/<model>, and duplicates are removed. Retired models return 422 with a replacement; unknown models return 400. Null or omitted means no model restriction. A model excluded by project policy can be stored, but is excluded from effective_allowed_models and rejected at request time.

Required array length: 1 - 100 elements
Required string length: 1 - 200
requests_per_minute
integer | null

Inference requests this key may make per minute. Null or omitted: no rate limit.

Required range: 1 <= x <= 10000

Response

The key's inference limits as they now stand

api_key_id
string<uuid>
required
project_id
string<uuid>
required
configured
boolean
required

Whether the key carries limits. False means every limit below is a default.

enforced
boolean
required

Whether the gateway enforces the key’s limits. If false, stored limits do not restrict requests.

allowed_models
string[] | null
required
requests_per_minute
integer | null
required
effective_allowed_models
string[] | null
required

The key’s allowed_models after applying the project policy’s model restrictions. Other constraints, such as hosts, jurisdictions, and tier, apply to each request. Null if the key has no model list.

created_at
string | null
required
updated_at
string | null
required
updated_by
string<uuid> | null
required