API7 Docs
API7 GatewayAPI7 Enterprise Admin APIsList AI gateway span logs

List AI gateway span logs

GET /api/ai_gateway_logs

GET /api/ai_gateway_logs

Interactive request editor loads with JavaScript.

Authorization

X-API-KEY<token>

In: header

Query Parameters

ai_gateway_group_id?string

Filter logs by AI gateway group ID.

instance_id?string

Filter logs by dataplane instance ID.

model?string

Filter logs by model name.

start_time?string

Filter logs with start_time >= this value (RFC3339 format).

Formatdate-time
end_time?string

Filter logs with start_time < this value (RFC3339 format).

Formatdate-time
page?integer

Page number of the listed resources. Used together with page_size. For example, when there are 13 resources in total, if the query parameters are page=1&page_size=10, the GET response will show the route total as 13 and display 10 resources in the first page. If the query parameters are page=2&page_size=10, the GET response will show the route total as 13 and display 3 resources in the second page.

Formatint32
Range1 <= value
page_size?integer

Number of resources listed per page. Used together with page. For example, when there are 13 resources in total, if the query parameters are page=1&page_size=10, the GET response will show the route total as 13 and display 10 resources in the first page. If the query parameters are page=2&page_size=10, the GET response will show the route total as 13 and display 3 resources in the second page.

Formatint32
Range1 <= value

Response Body

application/json

application/json

curl -X GET "$API_BASE_URL/api/ai_gateway_logs"

A paginated list of AI gateway span logs.

{
  "list": [
    {
      "id": "string",
      "instance_id": "string",
      "ai_gateway_group_id": "string",
      "trace_id": "string",
      "span_id": "string",
      "key_id": "string",
      "model": "string",
      "input_tokens": 0,
      "output_tokens": 0,
      "cache_read_tokens": 0,
      "cache_write_tokens": 0,
      "duration_ms": 0,
      "status_code": 0,
      "start_time": "2019-08-24T14:15:22Z",
      "first_response_at": "2019-08-24T14:15:22Z",
      "created_at": "2019-08-24T14:15:22Z"
    }
  ],
  "total": 0
}
Complete operation details and schema variants

GET /api/ai_gateway_logs

List AI gateway span logs

IAM Action: ai_gateway:ListAIGatewayLogs, Resource: arn:api7:ai_gateway:aigatewaygroup/* Returns a paginated list of AI/LLM span logs extracted from OTLP traces sent by AI gateway dataplanes.

Parameters

  • ai_gateway_group_id (query, string, optional): Filter logs by AI gateway group ID.
  • instance_id (query, string, optional): Filter logs by dataplane instance ID.
  • model (query, string, optional): Filter logs by model name.
  • start_time (query, string, optional): Filter logs with start_time >= this value (RFC3339 format).
  • end_time (query, string, optional): Filter logs with start_time < this value (RFC3339 format).
  • page (query, integer, optional): Page number of the listed resources. Used together with page_size. For example, when there are 13 resources in total, if the query parameters are page=1&page_size=10, the GET response will show the route total as 13 and display 10 resources in the first page. If the query parameters are page=2&page_size=10, the GET response will show the route total as 13 and display 3 resources in the second page.
  • page_size (query, integer, optional): Number of resources listed per page. Used together with page. For example, when there are 13 resources in total, if the query parameters are page=1&page_size=10, the GET response will show the route total as 13 and display 10 resources in the first page. If the query parameters are page=2&page_size=10, the GET response will show the route total as 13 and display 3 resources in the second page.

Responses

  • 200: A paginated list of AI gateway span logs.
  • list (array, optional):
  • list[].id (string, optional): Unique log record ID (hex traceID + spanID).
  • list[].instance_id (string, optional): Dataplane instance ID from resource attribute service.instance.id.
  • list[].ai_gateway_group_id (string, optional): AI gateway group ID.
  • list[].trace_id (string, optional): OTLP trace ID (hex-encoded).
  • list[].span_id (string, optional): OTLP span ID (hex-encoded).
  • list[].key_id (string, optional): API key ID used for this request (aisix.apikey_id attribute).
  • list[].model (string, optional): LLM model name (gen_ai.request.model or gen_ai.response.model).
  • list[].input_tokens (integer, optional): Number of input tokens (gen_ai.usage.input_tokens).
  • list[].output_tokens (integer, optional): Number of output tokens (gen_ai.usage.output_tokens).
  • list[].cache_read_tokens (integer, optional): Number of cache-read input tokens.
  • list[].cache_write_tokens (integer, optional): Number of cache-creation input tokens.
  • list[].duration_ms (integer, optional): Total request duration in milliseconds.
  • list[].status_code (integer, optional): HTTP status code of the LLM response.
  • list[].start_time (string, optional): Request start time (span StartTimeUnixNano).
  • list[].first_response_at (string, optional): Time of first response token, derived from the first-token-arrived span event ("first token arrived" / kind=first_token_arrived).
  • list[].created_at (string, optional): Record creation timestamp.
  • total (integer, optional): Total number of matching log records.
  • 400: Bad Request — The request was malformed or contained invalid parameters. Check the error_msg field for details.
  • value (object, optional): The specific resource.
  • error_msg (string, optional): The error message.

cURL

curl -X GET '$API_BASE_URL/api/ai_gateway_logs'

AI Provider

Previous Page

Overview

Next Page