Skip to content

Commit b48a15c

Browse files
markmccomaniac
andcommitted
[V1][Metrics] Tweak prefix caching note
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com> Signed-off-by: Mark McLoughlin <markmc@redhat.com>
1 parent 674edca commit b48a15c

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

docs/source/design/v1/metrics.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -279,7 +279,7 @@ every 5 seconds with some key metrics:
279279
seconds
280280
- The number of new tokens generated per second over the past 5
281281
seconds
282-
- The prefix cache hit rate over the most recent 1 queries
282+
- The prefix cache hit rate over the most recent 1k kv-cache block queries
283283

284284
### Metrics Publishing - Prometheus
285285

0 commit comments

Comments
 (0)