Repository navigation
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
MI355X jobs that select the native SGLang client currently fall through to InferenceX, and runtime environment overrides are ignored. This adds an opt-in native mode for fixed random-ids workloads, plus a server launch wrapper that applies unset/JSON environment values after hardware defaults.
The native client runs full-output warmup before measurement, supports a minimum warmup count, and retains raw results and protocol logs. Only complete, fixed-length successful requests are published. Recognized compilation messages reject the interval without discarding its evidence or selecting a faster repeat. The default InferenceX and profiling paths remain unchanged.
Validation: targeted subprocess tests cover CLI arguments, warmup ordering, environment isolation, incomplete/truncated requests, native failures, compilation messages, generic dispatch, and helper packaging. No GPU throughput claim is made by these tests.