Skip to content

Comments

fix: tp weight sync#1292

Open
akoumpa wants to merge 20 commits intomainfrom
akoumparouli/fix_tp_weight_sync
Open

fix: tp weight sync#1292
akoumpa wants to merge 20 commits intomainfrom
akoumparouli/fix_tp_weight_sync

Conversation

@akoumpa
Copy link
Contributor

@akoumpa akoumpa commented Feb 15, 2026

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Changelog

  • Add specific line by line info of high level changes in this PR.

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you add or update any necessary documentation?

If you haven't finished some of the above items you can still open "Draft" PR.

Additional Information

  • Related to # (issue)

@akoumpa
fix tp weight
c60b1ba 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@copy-pr-bot
Copy link

copy-pr-bot bot commented Feb 15, 2026

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@akoumpa
Copy link
Contributor Author

akoumpa commented Feb 15, 2026

/ok to test c60b1ba

@akoumpa akoumpa linked an issue Feb 15, 2026 that may be closed by this pull request
Custom module weights not synchronized across TP ranks #1291
Open
@akoumpa
add yaml tp
979008c 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
lint
191350f 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
Copy link
Contributor Author

akoumpa commented Feb 15, 2026

/ok to test 191350f

@akoumpa
fix
f56bf3c 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
Copy link
Contributor Author

akoumpa commented Feb 15, 2026

/ok to test f56bf3c

@akoumpa
move fusedcolwise
4537708 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
update import
6dac5b4 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
move more
686cd55 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
test
13994ba 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
switch to vocabpar
ff18ed9 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
bump base container
9b40e9d 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa akoumpa requested a review from a team as a code owner February 19, 2026 07:23
@akoumpa
Copy link
Contributor Author

akoumpa commented Feb 19, 2026

/ok to test 9b40e9d

@akoumpa
ruff
14af195 
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
@akoumpa
Copy link
Contributor Author

akoumpa commented Feb 19, 2026

/ok to test 14af195

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

Reviewers

@ZhiyuLi-Nvidia ZhiyuLi-Nvidia ZhiyuLi-Nvidia left review comments

@HuiyingLi HuiyingLi Awaiting requested review from HuiyingLi HuiyingLi is a code owner

@adil-a adil-a Awaiting requested review from adil-a adil-a is a code owner

@hemildesai hemildesai Awaiting requested review from hemildesai hemildesai is a code owner

At least 1 approving review is required to merge this pull request.

Assignees

No one assigned

Labels

r0.3.0 Add for cherry-pick into release branch r0.3.0

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Custom module weights not synchronized across TP ranks

2 participants

@akoumpa @ZhiyuLi-Nvidia