genai_get_pricing

You can use this function to display pricing information, create cost calculators, audit pricing data, or understand the cost structure of different AI models.

Usage#

Syntax#

genai_get_pricing(model)

Parameters#

Name	Type	Required	Description
model	string	Yes	The name of the AI model (for example, 'gpt-4', 'claude-3-opus-20240229', 'gpt-3.5-turbo').

Returns#

Returns a dynamic object containing pricing information with the following structure:

{
  "input_price_per_million": <number>,
  "output_price_per_million": <number>
}

Returns null if the model isn't recognized.

Example#

Get pricing information for a GenAI model to understand cost structure.

Query

['otel-demo-genai']
| extend model = ['attributes.gen_ai.request.model']
| extend pricing = genai_get_pricing(model)
| where isnotnull(pricing)
| summarize request_count = count() by model, pricing = tostring(pricing)
| top 3 by request_count

Run in Playground

Output

model	pricing	request_count
gpt-4	`{"input_price_per_million": 30.0, "output_price_per_million": 60.0}`	450
gpt-3.5-turbo	`{"input_price_per_million": 0.5, "output_price_per_million": 1.5}`	1250

This query shows which models are being used and their associated pricing, helping teams make informed decisions.

genai_cost: Calculates total cost for a conversation. Use this for actual cost calculation after retrieving pricing information.
genai_input_cost: Calculates input token cost. This function uses genai_get_pricing internally to determine input costs.
genai_output_cost: Calculates output token cost. This function uses genai_get_pricing internally to determine output costs.
genai_estimate_tokens: Estimates token count. Combine with pricing information to predict costs before making API calls.

Other query languages#

Splunk SPL users

In Splunk SPL, you would typically use a lookup table to retrieve pricing information by model name.

Splunk example

| lookup model_pricing model OUTPUT input_price output_price

APL equivalent

['ai-logs']
| extend pricing = genai_get_pricing(model)

ANSI SQL users

In ANSI SQL, you would join with a pricing table to get model costs.

SQL example

SELECT
  l.*,
  p.input_price,
  p.output_price
FROM ai_logs l
JOIN model_pricing p ON l.model = p.model_name