Repository navigation
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
6 tasks done
caduri
marked this pull request as ready for review
October 4, 2026 11:18
This branch was successfully deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Documents the new
fire_and_forgetandfire_and_forget_max_inflightoptions for the generic guardrail API, added in BerriAI/litellm#43681. Please merge it only after that PR shipsThe page gains both options in the LiteLLM Configuration example and a short "Fire and forget" section with a warning that the guardrail becomes observe-only and can no longer block. The section explains how dispatched calls are recorded (
success, so they count as passes in the usage dashboard and compliance checks), that background failures are only logged, that a stream is checked once when it closes, including one the client hangs up on, the per-worker in-flight cap and itsnot_rundrops, the background timeout, and what an invalid value doesnpm run build,npm run lint:writingandnpm run lint:docsall pass locally