docs(prometheus): clarify what each request-latency metric measures - #991
Merged
Conversation
The Request Latency Metrics table jammed each metric's description and full label list into one dense run-on cell, and described litellm_request_total_latency_metric as "total latency... for a request" without saying it previously excluded authentication time (see the companion litellm fix). Rewritten with a request-lifecycle timeline, a "which metric should I alert on" guide, and label lists split out from descriptions.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
yucheng-berri
approved these changes
Aug 22, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
The Request Latency Metrics section of the Prometheus docs packed each metric's description and full label list into one dense table cell, and described
litellm_request_total_latency_metricas "total latency... for a request" without noting that it previously excluded authentication time.Rewrites the section around the user's task: choose the right metric, understand its exact measurement window, and query end-to-end p95 latency. The main content is now one comparison table, while the long label inventory is available in an expandable reference. Wording matches the corrected metric behavior landed in BerriAI/litellm#37958.
Linear ticket
Resolves LIT-6013
Test plan
node scripts/check-writing-style.js docs/proxy/prometheus.md: clean, no em dashesnpm run build: succeeds; the 11 pre-existing SSG warnings are all on unrelated pages (contact, knowledgebase, mcp, configs, email, logging, google_adk, several release notes), none on this page/docs/proxy/prometheus#request-latency-metricslocally and verified the comparison table, PromQL example, and expanded label reference