Metering API

Metering service API Documentation

---

Version: 1.2.0

Paths

Healthcheck API -

/aiwb/metering/api/v1/healthz

• Method: GET
• Summary: Health Check
• Description: Return Health Status. :return:
• Operation ID: health_check_aiwb_metering_api_v1_healthz_get
• Responses:
  • 200:
    • Description: Successful Response
    • Content-Type: application/json

Create Metering -

Endpoint: /aiwb/metering/api/v1/

• Method: POST

• Summary: Create Metering To Analyze Token Consumption
• Description: This API facilitates the logging of token consumption by tenants, users, and bots within the system. It allows for detailed tracking and analysis of token usage patterns, enabling administrators to monitor resource allocation efficiently. The endpoint supports specifying the relevant tenant, user, and bot identities for precise consumption tracking and auditing purposes.
• Operation ID: Create_Metering_to_analyze_token_consumption_aiwb_metering_api_v1__post

Request Payload:

Type	Parameter	Description	Data Type	is_optional
Query	bot_id	Identifier of the chatbot	string	madatory
Query	application_name	Name of the application using the chatbot	string	Optional
Query	unique_query_id	ID for the query	string	Optional
Query	user_query	User's input query	string	madatory
Query	model_answer	Answer generated by the chatbot model	string	Optional
Query	model_name	Name of the chatbot model used (OpenAI, Palm)	string	Optional
Query	pre_calc_input_token	Number of input tokens pre-calculated	integer	Optional
Query	input_token	Number of input tokens processed	integer	madatory
Query	output_token	Number of output tokens generated	integer	madatory
Query	input_payload	Additional input payload for the chatbot	string	madatory
Query	model_response	Full response generated by the chatbot model	string	madatory
Query	latency_in_seconds	Response latency in seconds	number	Optional
Query	word_count_sent	Number of words in the user query	integer	Optional
Query	word_count_received	Number of words in the chatbot's response	integer	Optional
Query	toxicity	Toxicity level of the response	string	Optional
Query	sentiment	Sentiment analysis result of the response	string	Optional

sample code

import requests
URL = "https://api.kadal.ai/aiwb/metering/api/v1/"

# sample python code
headers = { "Authorization ": "Bearer 1234567890" }

params:{
  "bot_id": "string",
  "application_name": "string",
  "unique_query_id": "string",
  "user_query": "string",
  "model_answer": "string",
  "model_type": "string",
  "model_name": "string",
  "pre_calc_input_token": 0,
  "input_token": 0,
  "output_token": 0,
  "metering_costs": {
    "input_token_cost": 0,
    "output_token_cost": 0,
    "av_translation_cost": 0,
    "embedding_cost": 0,
    "total_cost": 0
  },
  "input_payload": "string",
  "model_response": "string",
  "metering_logs": {
    "av_translation_duration": 0,
    "image_generation_nos": 0,
    "image_generation_quality": 0,
    "image_generation_dimension": "string"
  },
  "latency_in_seconds": 0,
  "word_count_sent": 0,
  "word_count_received": 0,
  "toxicity": "Toxic",
  "sentiment": "Positive"
}

response = requests.post(URL, headers=headers,params=params)
print(response)

Example Response value -

{
  "status_code": 201,
  "message": "Token consumption created successfully"
}

Responses:

Status Code	Description	Content-Type
200	Successful Response	application/json
201	Token consumption created successfully	application/json
400	Bad Request	application/json
500	Internal Server Error	application/json
422	Validation Error	application/json

Token Consumption by Tenant -

Endpoint: /aiwb/metering/api/v1/token_by_tenant/{tenant_id}

• Method: GET

• Summary: Retrieve the Usage Details of Tokens Consumed by a Specific Tenant
• Description: This API endpoint enables retrieval of token consumption data associated with a specific Tenant ID. By making a GET request to this endpoint with the relevant Tenant ID specified as a parameter, you can obtain detailed information about token usage within the specified tenant.
• Operation ID: Retrieve_the_usage_details_of_tokens_consumed_by_a_specific_tenant_aiwb_metering_api_v1_token_by_tenant__tenant_id__get

Request Payload:

Type	Parameter	Description	Data Type	is_optional
parameter	tenant_id	Tenant ID to retrive metering	string (UUID)	mandatory

Sample code

import requests
URL = "https://api.kadal.ai/aiwb/metering/api/v1/token_by_tenant/{tenant_id}"
# sample python code
headers = { "Authorization ": "Bearer 1234567890" }

params:{
  "tenant_id":"string"
}
response = requests.get(URL, headers=headers,params=params)
print(response)

Example Response value -

{
  "status_code": 200,
  "message": "List of token consumed by Tenant Id"
}

Responses:

Status Code	Description	Content-Type
200	List Token consumed by Tenant ID	application/json
404	Token consumption not found for given tenant	application/json
400	Bad Request	application/json
500	Internal Server Error	application/json
422	Validation Error	application/json

Token Consumption by User -

Endpoint: /aiwb/metering/api/v1/token_by_user/{user_id}

• Method: GET

-- Summary: Retrieve the Usage Details of Tokens Consumed by a Specific User

• Description: This API allows you to retrieve a list of tokens consumed by a specific User. By sending a GET request to this endpoint with the UserId as a parameter, you can access detailed information about token consumption associated with that user.
• Operation ID: Retrieve_the_usage_details_of_tokens_consumed_by_a_specific_user_aiwb_metering_api_v1_token_by_user__user_id__get

Request Payload:

Type	Parameter	Description	Data Type	is_optional
parameter	user_id	user ID to retrive metering	string (UUID)	mandatory

Sample code

import requests
URL = "https://api.kadal.ai/aiwb/metering/api/v1/token_by_user/{user_id}"
# sample python code
headers = { "Authorization ": "Bearer 1234567890" }

params:{
  "user_id":"string"
}
response = requests.get(URL, headers=headers,params=params)
print(response)

Example Response value -

{
  "status_code": 200,
  "message": "List of token consumed by UserId"
}

Responses:

Status Code	Description	Content-Type
200	List of Tokens consumed by UserId	application/json
404	Token consumption not found for given user	application/json
400	Bad Request	application/json
500	Internal Server Error	application/json
422	Validation Error	application/json

Components

Schemas

HTTPValidationError

• Type: object
• Description:
• Properties:
  • detail:
    • Type: array

ListTokenConsumedResponse

• Type: object
• Description:
• Properties:
  • status_code:
    • Type: integer
  • message:
    • Type: string
  • data:

MeteringResponse

• Type: object
• Description:
• Properties:
  • status_code:
    • Type: integer
  • message:
    • Type: string

Msg

• Type: object
• Description:
• Properties:
  • msg:
    • Type: string

Sentiment

• Type: Not specified
• Description: An enumeration.
• Properties:

TokenConsumption

• Type: object
• Description:
• Properties:
  • bot_id:
    • Type: string
  • application_name:
    • Type: string
  • unique_query_id:
    • Type: string
  • user_query:
    • Type: string
  • model_answer:
    • Type: string
  • model_name:
    • Type: string
  • pre_calc_input_token:
    • Type: integer
  • input_token:
    • Type: integer
  • output_token:
    • Type: integer
  • input_payload:
  • model_response:
  • latency_in_seconds:
    • Type: number
  • word_count_sent:
    • Type: integer
  • word_count_received:
    • Type: integer
  • toxicity:
  • sentiment:

Toxicity

• Type: Not specified
• Description: An enumeration.
• Properties:

ValidationError

• Type: object
• Description:
• Properties:
  • loc:
    • Type: array
  • msg:
    • Type: string
  • type:
    • Type: string