Early Termination of a Request

Currently, this API supports only early termination of Triton inference requests.

Function

Terminates a request in advance based on the Triton API definition.

Format

Operation type: POST

URL: https://{ip}:{port}/v2/models/${MODEL_NAME}[/versions/${MODEL_VERSION}]/stopInfer

  • Replace {ip} and {port} with the IP address and port number of the service plane, that is, ipAddress and port.
  • The ${MODEL_NAME} field specifies the name of the model to be queried.
  • The [/versions/${MODEL_VERSION}] field is not supported currently and is not passed.

Request Parameters

Parameter

Mandatory/Optional

Description

Value

id

Mandatory

Inference request ID

The value is a string of a maximum of 256 characters.

Usage Example

Request example:

POST https://{ip}:{port}/v2/models/llama3-70b/stopInfer

Request body:

{
    "id":"a123"
}

Response example:

{
    "id":"a123"
}

Response status code: 200

Output Description

Return Value

Type

Description

id

String

ID of the inference request that is successfully terminated.

In 2.1.RC1 and earlier versions, after this API is called, the inference is not terminated immediately. Instead, the inference is processed together with scheduling in the executor, and an invalid token is returned. In 2.2.RC1 and later versions, the inference is terminated immediately after this API is called, and no invalid token is returned.