Name and Version

version: 6529 (1eeb523)
built with cc (Ubuntu 11.4.0-1ubuntu1~22.04.2) 11.4.0 for x86_64-linux-gnu

Operating systems

Linux

Which llama.cpp modules do you know to be affected?

llama-server

Command line

Problem description & steps to reproduce

(Not exactly a bug, but it is also not a FR because this feature was already available before)

In the previous WebUI pp/tg stats were visible when hovering over the always visible "Speed: X t/s" part of a reply like this:

This allowed easy comparison of these stats when adjusting the parameters of a model or when comparing different models. I was able to regenerate a reply with a new model and see the quality difference alongside the processing speed impact.

In the new UI this information disappears after the generation. So please bring this useful part of the ui back. If clutter is a problem, this can be visible only if the user enables the "Show tokens per second" setting.

(If I understand correctly these stats are still calculated and stored, because I downgraded to the previous version of WebUI and still can see the stats from the new UI.)

First Bad Commit

No response

Relevant log output