Skip to content

Update Qwen3.5 VL Long Context Recipe Default Configs - #6470

Merged
zhongbozhu merged 3 commits into
NVIDIA-NeMo:mainfrom
zhongbozhu:qwen35-long-context-defaults
Oct 17, 2026
Merged

zhongbozhu merged 3 commits into
NVIDIA-NeMo:mainfrom
zhongbozhu:qwen35-long-context-defaults

Conversation

@zhongbozhu

Copy link
Copy Markdown
Contributor

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.

GitHub Actions CI

See the CI section in the Contributing doc for how to trigger the CI. A Nvidia developer will need to approve and trigger the CI for external contributors.

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?
  • Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
    • Reviewer: Does the PR have correct import guards for all optional libraries?

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

Additional Information

  • Related to # (issue)

Signed-off-by: zhongboz <zhongboz@nvidia.com>
@copy-pr-bot

copy-pr-bot Bot commented Oct 14, 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.

@zhongbozhu
zhongbozhu marked this pull request as ready for review October 14, 2026 23:36
@zhongbozhu
zhongbozhu requested a review from cuichenx October 14, 2026 23:36
@HanChennn
HanChennn force-pushed the qwen35-long-context-defaults branch from 344db94 to ae438ac Compare October 15, 2026 00:54
Signed-off-by: zhongboz <zhongboz@nvidia.com>
@zhongbozhu
zhongbozhu force-pushed the qwen35-long-context-defaults branch from ae438ac to 79bc218 Compare October 15, 2026 01:30
Select the Transformer Engine/cuDNN GDN backend and disable DP
gradient and parameter overlap in the GB200 text and VL long-context
SFT recipes. Use one validation worker for the VL Energon dataset.

Extend real-provider and runtime-overlap checks for both precision
variants, including the pure-text vision-field and BF16 boundaries.

Signed-off-by: HanChennn <chankchen.02@gmail.com>
@cuichenx cuichenx added feature New capabilities, enhancements, or enablement work area:recipe Training recipes and launch configs needs-review PR is ready for code review and waiting on a reviewer labels Oct 15, 2026
@cuichenx

Copy link
Copy Markdown
Contributor

/ok to test fe05108

@github-actions

Copy link
Copy Markdown
Contributor

Automatic Claude reviews have been retired. To request a pull-request review, post a comment containing:

/review

Add model=claude to use a Claude reviewer (the default is model=codex). mode=light|strict selects the review depth; for example, /review model=claude mode=strict. Comment /review help for all options.

@github-actions

Copy link
Copy Markdown
Contributor

@cuichenx cuichenx added ready-to-merge PR is approved, current, and only waiting for CI to pass before merge and removed needs-review PR is ready for code review and waiting on a reviewer labels Oct 16, 2026
@zhongbozhu
zhongbozhu merged commit 20dd692 into NVIDIA-NeMo:main Oct 17, 2026
95 checks passed

This branch was successfully deployed

1 active deployment
test — fe05108c Deployed Oct 16, 2026 by copy-pr-bot[bot] via cicd-wait-in-queue #22924
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area:recipe Training recipes and launch configs feature New capabilities, enhancements, or enablement work ready-to-merge PR is approved, current, and only waiting for CI to pass before merge

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants