Jump to

• Request
• Response

(Preview) Search model deployment inferences

post

/20260430/aiDataPlatforms/{aiDataPlatformId}/mlops/api/2.0/model-deployments/{deploymentId}/monitoring/inferences/search

(Preview) Returns paginated request and response-correlated inference invocation records for a model deployment. Each response contains at most 20 rows, OCI Logging pagination is used internally, and only the first 10,000 matching rows can be retrieved within a maximum 14-day range. When no time range is provided, the previous 24 hours is used.

Request

Path Parameters

• aiDataPlatformId(required): string
  Minimum Length: 1

  Maximum Length: 500

  The [OCID](/iaas/Content/General/Concepts/identifiers.htm) of the AI Data Platform (Data Lake) instance.
• deploymentId(required): string
  Minimum Length: 32

  Maximum Length: 32

  Pattern: ^[0-9a-f]{32}$

  The unique 32-character hexadecimal ID of the model deployment.

Header Parameters

• opc-request-id: string
  Maximum Length: 255

  Unique Oracle-assigned identifier for the request. If you need to contact Oracle about a particular request, please provide the request ID. The only valid characters for request IDs are letters, numbers, underscore, and dash.

Body ()

Timestamp range, optional post-correlation filters, and pagination settings.

Root Schema : InferenceInvocationSearchDetails

Type: object

Filters and OCI-backed pagination for correlated inference invocation rows. Each response contains at most 20 rows; only the first 10,000 matching rows are available.

Show Source

• modelVersion: string
  Minimum Length: 1

  Maximum Length: 64

  Restricts results to invocations served by this model version.
• pageToken: string
  Minimum Length: 1

  Opaque token from a previous response. It continues pagination of the filtered, correlated invocation rows.
• requestUri: string
  Minimum Length: 1

  Maximum Length: 2048

  Restricts results to invocations against this inference endpoint URI.
• status: string
  Allowed Values: [ "SUCCESS", "MALFORMED", "FAILED", "UNKNOWN" ]

  Restricts results to correlated invocations with this status.
• timeEnd: string(date-time)
  Inclusive request timestamp end. Defaults to current time when omitted. The range must be within the most recent 14 days. Body timestamps require millisecond precision. Example: 2026-09-15T00:00:00.000Z.
• timeStart: string(date-time)
  Inclusive request timestamp start. Defaults to current time minus 24 hours when omitted. The range must be within the most recent 14 days. Body timestamps require millisecond precision. Example: 2026-09-01T00:00:00.000Z.

{
    "description":"Filters and OCI-backed pagination for correlated inference invocation rows. Each response contains at most 20 rows; only the first 10,000 matching rows are available.",
    "type":"object",
    "properties":{
        "timeStart":{
            "type":"string",
            "format":"date-time",
            "description":"Inclusive request timestamp start. Defaults to current time minus 24 hours when omitted. The range must be within the most recent 14 days. Body timestamps require millisecond precision. Example: 2026-09-01T00:00:00.000Z."
        },
        "timeEnd":{
            "type":"string",
            "format":"date-time",
            "description":"Inclusive request timestamp end. Defaults to current time when omitted. The range must be within the most recent 14 days. Body timestamps require millisecond precision. Example: 2026-09-15T00:00:00.000Z."
        },
        "status":{
            "description":"Restricts results to correlated invocations with this status.",
            "type":"string",
            "x-obmcs-top-level-enum":"#/definitions/InferenceInvocationStatus",
            "enum":[
                "SUCCESS",
                "MALFORMED",
                "FAILED",
                "UNKNOWN"
            ]
        },
        "modelVersion":{
            "description":"Restricts results to invocations served by this model version.",
            "type":"string",
            "minLength":1,
            "maxLength":64
        },
        "requestUri":{
            "description":"Restricts results to invocations against this inference endpoint URI.",
            "type":"string",
            "minLength":1,
            "maxLength":2048
        },
        "pageToken":{
            "description":"Opaque token from a previous response. It continues pagination of the filtered, correlated invocation rows.",
            "type":"string",
            "minLength":1
        }
    }
}

Back to Top

Response

Supported Media Types

• application/json

200 Response

Successful operation. Correlated inference invocation records are returned.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : InferenceInvocationSearchResult

Type: object

Paginated correlated inference invocation rows. Each response contains at most 20 rows; pagination is backed by OCI Logging. At most 10,000 matching rows can be retrieved.

Show Source

• items(required): array items
  Maximum Number of Items: 20

  Correlated inference invocation rows for the requested page.
• modelVersions(required): array modelVersions
  Model versions configured on the requested deployment.
• nextPageToken: string
  Opaque token for retrieving the next page. Omitted when no additional page is available.
• requestUris(required): array requestUris
  Available filter values: the distinct inference endpoint URIs found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.
• statuses(required): array statuses
  Available filter values: the distinct invocation statuses found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.
• timeEnd(required): string(date-time)
  Inclusive end time resolved for this search query.
• timeStart(required): string(date-time)
  Inclusive start time resolved for this search query.

{
    "description":"Paginated correlated inference invocation rows. Each response contains at most 20 rows; pagination is backed by OCI Logging. At most 10,000 matching rows can be retrieved.",
    "type":"object",
    "required":[
        "timeStart",
        "timeEnd",
        "items",
        "statuses",
        "requestUris",
        "modelVersions"
    ],
    "properties":{
        "timeStart":{
            "description":"Inclusive start time resolved for this search query.",
            "type":"string",
            "format":"date-time"
        },
        "timeEnd":{
            "description":"Inclusive end time resolved for this search query.",
            "type":"string",
            "format":"date-time"
        },
        "items":{
            "description":"Correlated inference invocation rows for the requested page.",
            "type":"array",
            "maxItems":20,
            "items":{
                "$ref":"#/definitions/InferenceInvocationSummary"
            }
        },
        "nextPageToken":{
            "description":"Opaque token for retrieving the next page. Omitted when no additional page is available.",
            "type":"string"
        },
        "statuses":{
            "description":"Available filter values: the distinct invocation statuses found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.",
            "type":"array",
            "items":{
                "$ref":"#/definitions/InferenceInvocationStatus"
            }
        },
        "requestUris":{
            "description":"Available filter values: the distinct inference endpoint URIs found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.",
            "type":"array",
            "items":{
                "type":"string"
            }
        },
        "modelVersions":{
            "description":"Model versions configured on the requested deployment.",
            "type":"array",
            "items":{
                "type":"string"
            }
        }
    }
}

Nested Schema : items

Type: array

Maximum Number of Items: 20

Correlated inference invocation rows for the requested page.

Show Source

• Array of: object InferenceInvocationSummary
  One correlated inference request and response record.

{
    "description":"Correlated inference invocation rows for the requested page.",
    "type":"array",
    "maxItems":20,
    "items":{
        "$ref":"#/definitions/InferenceInvocationSummary"
    }
}

Nested Schema : modelVersions

Type: array

Model versions configured on the requested deployment.

Show Source

• Array of: string

{
    "description":"Model versions configured on the requested deployment.",
    "type":"array",
    "items":{
        "type":"string"
    }
}

Nested Schema : requestUris

Type: array

Available filter values: the distinct inference endpoint URIs found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.

Show Source

• Array of: string

{
    "description":"Available filter values: the distinct inference endpoint URIs found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.",
    "type":"array",
    "items":{
        "type":"string"
    }
}

Nested Schema : statuses

Type: array

Available filter values: the distinct invocation statuses found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.

Show Source

• Array of: string
  Allowed Values: [ "SUCCESS", "MALFORMED", "FAILED", "UNKNOWN" ]

  Status derived from the correlated inference response event. UNKNOWN is reported when the response event carries no status code, which some model runtime versions omit; the invocation is still returned rather than discarded.

{
    "description":"Available filter values: the distinct invocation statuses found across the whole requested time range. This is not narrowed by the filters in the request, so it stays stable while a caller changes filters.",
    "type":"array",
    "items":{
        "$ref":"#/definitions/InferenceInvocationStatus"
    }
}

Nested Schema : InferenceInvocationSummary

Type: object

One correlated inference request and response record.

Show Source

• latencyMs: integer(int64)
  Minimum Value: 0

  Response latency in milliseconds.
• modelVersion: string
  Model version associated with the invocation.
• requestId(required): string
  Request ID of the inference invocation, used to correlate the request and response events. Pass this value to the get inference details operation.
• requestTimestamp(required): string(date-time)
  Timestamp recorded for the inference request event.
• requestUri(required): string
  Inference endpoint URI.
• status(required): string
  Allowed Values: [ "SUCCESS", "MALFORMED", "FAILED", "UNKNOWN" ]

  Status derived from the correlated inference response event.

{
    "description":"One correlated inference request and response record.",
    "type":"object",
    "required":[
        "requestTimestamp",
        "requestUri",
        "status",
        "requestId"
    ],
    "properties":{
        "requestTimestamp":{
            "description":"Timestamp recorded for the inference request event.",
            "type":"string",
            "format":"date-time"
        },
        "requestUri":{
            "description":"Inference endpoint URI.",
            "type":"string"
        },
        "modelVersion":{
            "description":"Model version associated with the invocation.",
            "type":"string"
        },
        "status":{
            "description":"Status derived from the correlated inference response event.",
            "type":"string",
            "x-obmcs-top-level-enum":"#/definitions/InferenceInvocationStatus",
            "enum":[
                "SUCCESS",
                "MALFORMED",
                "FAILED",
                "UNKNOWN"
            ]
        },
        "latencyMs":{
            "description":"Response latency in milliseconds.",
            "type":"integer",
            "format":"int64",
            "minimum":0
        },
        "requestId":{
            "description":"Request ID of the inference invocation, used to correlate the request and response events. Pass this value to the get inference details operation.",
            "type":"string"
        }
    }
}

400 Response

Bad Request (invalid query parameters, malformed headers, and so on).

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

401 Response

Unauthorized (missing or expired credentials, and so on).

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

404 Response

Not Found. The requested resource was not found.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

429 Response

Too Many Requests. Too many requests sent to the server in a short period.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

500 Response

Internal Server Error. The server encountered an unexpected condition preventing fulfilment of the request.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

Default Response

Unknown Error. Error is not recognized by the system.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

Back to Top