Start Agent conversation (OpenAI protocol)

POST 
/v1/aibot/openai/chat/completions

Start Agent conversation (OpenAI protocol)

Request

application/json

Body

required

model stringrequired

Agent ID

messages object[]required
Messages
Array [
role stringrequired
Role in the conversation history
content stringrequired
Content of the conversation history
]
stream booleanrequired

Whether to return streaming responses

Responses

x-stream:false

Returned when the Agent conversation succeeds with non-streaming response

Response Headers

• X-Request-Id string

  Request ID

application/json

Schema

Schema

id string

Conversation record ID

object string

API type

created number

Conversation timestamp

model string

Agent ID

choices undefined[]
choices
Array [
index number
index
message object
role string
Message role
content string
Message content
finish_reason string
Conversation termination flag; continue means not finished, stop means finished
]
usage object

Token usage

prompt_tokens number

Number of prompt tokens; remains constant across multiple responses

completion_tokens number

Total completion tokens. In streaming responses, this accumulates across chunks.

total_tokens number

Sum of prompt_tokens and completion_tokens

Example (from schema)

{
  "id": "record-427d3652",
  "object": "chat.completion",
  "created": "1726030224",
  "model": "bot-xxxx",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello"
      },
      "finish_reason": "continue"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 20,
    "total_tokens": 0
  }
}

x-stream:true

Returned when the Agent conversation succeeds with streaming response

application/json

Schema

Schema

id string

Conversation record ID

object string

API type

created number

Conversation timestamp

model string

Agent ID

choices undefined[]
choices
Array [
index number
index
delta object
role string
Message role
content string
Message content
finish_reason string
Conversation termination flag; continue means not finished, stop means finished
]
usage object

Token usage

prompt_tokens number

Number of prompt tokens; remains constant across multiple responses

completion_tokens number

Total completion tokens. In streaming responses, this accumulates across chunks.

total_tokens number

Sum of prompt_tokens and completion_tokens

Example (from schema)

{
  "id": "record-427d3652",
  "object": "chat.completion",
  "created": "1726030224",
  "model": "bot-xxxx",
  "choices": [
    {
      "index": 0,
      "delta": {
        "role": "assistant",
        "content": "Hello"
      },
      "finish_reason": "continue"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 20,
    "total_tokens": 0
  }
}

default

Returned when the Agent conversation fails

application/json

Schema

Schema

code string

Error code

message string

Error message

requestId string

Request ID

Example (from schema)

{
  "code": "INVALID_CREDENTIALS",
  "message": "Credentials are invalid",
  "requestId": "6360ba4b-f672-434b-9dee-35686a5589a4"
}

Loading...