Skip to content

cp: Merge main to super-v3.5-posttraining - 2026-08-03 - #3482

Open
yfw wants to merge 8 commits into
super-v3.5-posttrainingfrom
yifu/super-merge-main-20260803
Open

cp: Merge main to super-v3.5-posttraining - 2026-08-03#3482
yfw wants to merge 8 commits into
super-v3.5-posttrainingfrom
yifu/super-merge-main-20260803

Conversation

@yfw

@yfw yfw commented Aug 3, 2026

Copy link
Copy Markdown
Contributor

What does this PR do ?

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

Issues

List issues that this PR closes (syntax):

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this

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 run the unit tests and functional tests locally? Visit our Testing Guide for how to run tests
  • Did you add or update any necessary documentation? Visit our Document Development Guide for how to write, build and test the docs.

Additional Information

  • ...

zyzhou5 and others added 8 commits July 31, 2026 16:17
@zyzhou5 @claude
feat(sc): support router replay with TQ (#3378)
1afc767 
Signed-off-by: Zeyu Zhou <zezhou@nvidia.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
@yfw
feat(gym): add Ultra environment prefetch config (#3452)
c6b56da 
Signed-off-by: Yi-Fu Wu <yifu.wu@gmail.com>
@svcnemo-autobot
chore(ci): AUT-1231 bump preflight workflow template (#3460)
5529625 
Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@michal2409
feat(grpo): optional early stop at a validation metric threshold (#3404)
4dbfa20 
Signed-off-by: Michal Futrega <mfutrega@nvidia.com>
@hchings
fix(trtllm): Add TRTLLM SWE training example, update refit APIs (#3418)
8fb82ca 
Signed-off-by: Erin Ho <erinh@nvidia.com>
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
@RayenTian
fix(sc): log config parameters to W&B (#3465)
89f4d1f 
Signed-off-by: ruit <ruit@nvidia.com>
@tianyi-zhang-02 @claude
perf(loss): avoid full-vocab student log-softmax in same-vocab top-k …
5f5d82b 
…KD (#3314)

Signed-off-by: Tianyi Zhang <123608656+tianyi-zhang-02@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
@tianyi-zhang-02 @claude @yuki-97
fix(data): honor subset on GSM8KDataset and correct the val-split docs (
00d3fc2 
#3389)

Signed-off-by: Tianyi Zhang <123608656+tianyi-zhang-02@users.noreply.github.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Yuki Huang <yukih@nvidia.com>
@yfw
yfw requested review from a team and terrykong as code owners August 3, 2026 23:56
@copy-pr-bot

copy-pr-bot Bot commented Aug 3, 2026

Copy link
Copy Markdown

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.

@github-actions github-actions Bot added Documentation Improvements or additions to documentation CI Relating to CI labels Aug 4, 2026
@yfw yfw changed the title Merge main to super-v3.5-posttraining - 2026-08-03 cp: Merge main to super-v3.5-posttraining - 2026-08-03 Aug 4, 2026
@yfw yfw added the CI:Lfast Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version) label Aug 4, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

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

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

Assignees

No one assigned

Labels

CI:Lfast Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version) CI Relating to CI Documentation Improvements or additions to documentation

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.