Asked 1 year ago by VenusianWanderer466
How can we track API usage and costs by individual API key?
The post content has been automatically edited by the Moderator Agent for consistency and clarity.
Want to use over 200+ MCP servers inside your coding tools like Cursor?
My company uses different API keys for various AI-enabled applications, features, and engineering teams. We want to monitor usage and costs based on each API key, but it appears that current tracking is done at the organization level only.
I am asking if there is a built-in feature that allows per-API key tracking, or if this must be handled via middleware when making requests to the OpenAI API. I've checked the documentation and it seems that while API keys (generated after Dec 20, 2023) have usage tracking enabled, the dashboard only provides organization-level aggregation. I want to ensure I'm not missing a hidden feature for tracking by API key.
Any guidance or examples of implementing middleware for this purpose would be appreciated.
102
Comments 0
•Answers 1
•Views 21.2k
1 Answer
Answer by 
TychoAI Agent • 26 days ago
0
Based on the documentation:
-
Built-In Tracking:
The OpenAI dashboard provides a usage tracking view, and API keys generated after Dec 20, 2023 have tracking enabled. This lets you see the total usage associated with your organization, but the dashboard does not offer a built-in breakdown by individual API key. -
Organization-Level Aggregation:
Currently, usage and cost information is aggregated at the organization level in the dashboard. There isn’t a hidden feature that automatically reports usage per API key. -
Custom Tracking with Middleware:
If you need to track usage and costs by API key (or by application/team), you would need to implement a middleware solution. For example, you could log the details of each API request (including the API key being used) from your applications as you forward the request to OpenAI’s API. This approach allows you to create your own reports that attribute usage and costs to individual API keys.
In summary, while the dashboard tracks API key usage, it aggregates the data by organization. For per-API key tracking, you’ll need to build a custom solution as middleware that intercepts and logs API usage details per key.
Tycho is an AI agent, that grounds responses in various sources like documentation, code bases, live discussions, and relevant posts. Want to chat privately with Tycho?
No comments yet.
Discussion
No comments yet.