Skip to content

Added gb200 glm5 fp8 mtp recipes - #237

Open
nvjullin wants to merge 3 commits into
NVIDIA:mainfrom
nvjullin:gb200-glm5-fp8-mtp
Open

Added gb200 glm5 fp8 mtp recipes#237
nvjullin wants to merge 3 commits into
NVIDIA:mainfrom
nvjullin:gb200-glm5-fp8-mtp

Conversation

@nvjullin

@nvjullin nvjullin commented Jul 2, 2026

Copy link
Copy Markdown
Contributor

Few caveats:

  1. May need to update container to nightly for bug fix [PD] Fix data race in NixlKVManager for NIXL backend sgl-project/sglang#28897. If configs runs properly, bug is likely benign.
  2. 1k1k low-latency is suboptimal, using DTP8 instead of TP8. TP8 has bugs, fix in progress.

TP8 added, it was an intermittent failure only.

@codecov-commenter

codecov-commenter commented Jul 2, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (main@53fffaa). Learn more about missing BASE report.

Additional details and impacted files
@@           Coverage Diff           @@
##             main     #237   +/-   ##
=======================================
  Coverage        ?   65.93%           
=======================================
  Files           ?       67           
  Lines           ?     8436           
  Branches        ?        0           
=======================================
  Hits            ?     5562           
  Misses          ?     2874           
  Partials        ?        0           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants