Update GB300 FP4 GLM5 low-latency sweep#175
Open
weireweire wants to merge 1 commit into
Open
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #175 +/- ##
=======================================
Coverage ? 65.10%
=======================================
Files ? 67
Lines ? 8217
Branches ? 0
=======================================
Hits ? 5350
Misses ? 2867
Partials ? 0 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
weireweire
requested review from
hjjq,
kedarpotdar-nv,
kyleliang-nv and
qiching
as code owners
weireweire
force-pushed
the
gb300-fp4-glm5-lowlat-concurrency
branch
from
80952de to
41d73ff
Compare
functionstackx
added a commit
to SemiAnalysisAI/InferenceX
that referenced
this pull request
May 29, 2026
…pology (#1583) * glm5-fp4-gb300-dynamo-sglang: extend 8k1k low-lat sweep with 1p17d topology Mirrors NVIDIA/srt-slurm#175: adds a 5th 8k1k_stp_lowlat_4 recipe with decode_nodes/workers=17, and lowers per-zip-index decode max-running-requests / cuda-graph-max-bs from a flat 4096 to 128/64/32/16/1 across lowlat_0..4. Benchmark concurrencies follow suit: 128/64/32/16/12. nvidia-master.yaml conc-list updated to match for each of the five 1p{3,5,9,15,17}d entries. * perf-changelog: set PR link to #1583 --------- Co-authored-by: functionstackx <47992694+functionstackx@users.noreply.github.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Updates the GB300 FP4 GLM5 8k1k low-latency sweep to add a fifth decode point and align max-running-requests, CUDA graph batch sizes, and SA-Bench concurrencies.
Validation: