Skip to content

Use token usage from Vertex response

Nicolas Dular requested to merge nd/use-vertex-token-response into master

What does this MR do and why?

Related issue: #428178 (closed)

Instead of relying on an estimation of 4 characters being 1 token, we now use the actual tokens we receive from the Vertex API. In addition to that, we now track embeddings as a separate action and do not count for it twice for input and output.

Screenshots or screen recordings

Screenshots are required for UI changes, and strongly recommended for all other merge requests.

Before After

How to set up and validate locally

Numbered steps to set up and validate the change are strongly suggested.

MR acceptance checklist

This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.

Merge request reports