List LLM Requests

get

https://api.cast.ai/ai-optimizer/v1beta/organizations/{organizationId}/llm-requests

Lists individual LLM request entries (chat completion inference logs) with
pagination and filtering, for a log-viewer style read path.

Recent Requests

Time	Status	User Agent
Retrieving recent requests…

Loading…

Path Params

organizationId

string

required

The ID of the organization.

Query Params

startTime

date-time

Start of the time window. Defaults to 7 days ago. Bypassed when request_id is set.

endTime

date-time

End of the time window. Defaults to now. Bypassed when request_id is set.

page.limit

string

Defines the number of items that should be returned

page.cursor

string

Cursor that defines token indicating where to start the next page.
Empty value indicates to start from beginning of the dataset.

sort.field

string

The field name to order by. Nested fields can be specified by using a dot (e.g. user.name).

sort.order

string

enum

The order of returned items. Default is ASC.

Allowed:

models

array of strings

Filter by model names. Empty = all models.

models

providers

array of strings

Filter by provider types (e.g. "openai", "groq"). Empty = all providers.

providers

providerNames

array of strings

Filter by provider names (user-defined config aliases). Empty = all.

providerNames

agentSources

array of strings

Filter by agent sources (OTEL client identities, e.g. "claude-code", "pi"). Empty = all.

agentSources

castaiApiKeyIds

array of strings

Filter by CAST AI API key IDs. Empty = all.

castaiApiKeyIds

castaiApiKeyOwnerTypes

array of strings

Filter by API key owner types ("serviceAccount" or "user"). Empty = all.

castaiApiKeyOwnerTypes

castaiApiKeyOwnerIds

array of strings

Filter by API key owner IDs. Empty = all.

castaiApiKeyOwnerIds

teamIds

array of strings

Filter by team IDs. Empty = all.

teamIds

tags

tagMatchMode

string

enum

How multiple tags entries are combined. UNSPECIFIED behaves like ALL.

Allowed:

statuses

array of strings

Filter by response status class. Empty = all statuses.

statuses

Allowed:

minTotalPrice

double

Minimum total price (USD), inclusive.

maxTotalPrice

double

Maximum total price (USD), inclusive.

minTotalTokens

string

Minimum total tokens, inclusive.

maxTotalTokens

string

Maximum total tokens, inclusive.

minResponseDuration

string

^-?(?:0|[1-9][0-9]{0,11})(?:\.[0-9]{1,9})?s$

Minimum response duration, inclusive.

maxResponseDuration

string

^-?(?:0|[1-9][0-9]{0,11})(?:\.[0-9]{1,9})?s$

Maximum response duration, inclusive.

requestId

string

Exact request ID lookup. When set, the time window is bypassed entirely.

sessionId

string

Filter by session ID.

inferUserFromApiKey

boolean

When true, filters results by the API key owner ID from the auth token.

Responses

Language

Credentials

URL

Loading…

Response

Click Try It! to start a request and see the response here! Or choose an example:

application/json

200OK

defaultDefault error response