OpenAI has announced the release of its Usage API, which provides developers with insights into activity across the OpenAI API, allowing them to better track costs.
Developers can monitor token usage by the minute, hour, or day, and filter usage by invoice line item, API key, project ID, user ID, model, and more.
The Usage API also comes with a separate Costs endpoint that can be used to manage costs, optimize workflows, and make faster decisions.
The company did note that the information given by the Usage API may not align with the Costs because of differences in how they record usage and spend, and as a result, it recommends companies not rely on it for any financial purposes.
The Usage API is available now, and more information on how to use it can be found in the API documentation.