[GH-ISSUE #9705] Server side token usage static log #52850

Open
opened 2026-04-29 01:10:51 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @darrkz on GitHub (Mar 13, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9705

I use ollama server,
But now ollama return token usage to client, it not easy to static the token usage of all clients.
So is there a way to log the token usage?
Should add it later?

Originally created by @darrkz on GitHub (Mar 13, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9705 I use ollama server, But now ollama return token usage to client, it not easy to static the token usage of all clients. So is there a way to log the token usage? Should add it later?
GiteaMirror added the feature request label 2026-04-29 01:10:51 -05:00
Author
Owner

@geiseri commented on GitHub (Mar 27, 2025):

Maybe a Prometheus endpoint? It's not a static log but it would be nice for reporting. My issue is that I am not always sure that context high/low water marks are. It is also very handy to know when you are trying to tune system prompts.

<!-- gh-comment-id:2758288739 --> @geiseri commented on GitHub (Mar 27, 2025): Maybe a Prometheus endpoint? It's not a static log but it would be nice for reporting. My issue is that I am not always sure that context high/low water marks are. It is also very handy to know when you are trying to tune system prompts.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#52850