Skip to content

Conversation

@kamran-nvidia
Copy link

@kamran-nvidia kamran-nvidia commented Jan 29, 2026

What does this PR do ?

  • Adds Ministral 3 script Examples(examples/models/vlm/ministral3/)

Changelog

Added complete example directory with:

  • README.md - Overview and links to each script
  • conversion.sh - HF ↔ Megatron checkpoint conversion commands
  • sft.sh - Full-parameter supervised fine-tuning with TP/PP variants
  • peft.sh - LoRA fine-tuning with TP/PP variants
  • inference.sh - Inference with HF, imported, and exported checkpoints

GitHub Actions CI

See the CI sectionin 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)

@copy-pr-bot
Copy link

copy-pr-bot bot commented Jan 29, 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.

@kamran-nvidia
Copy link
Author

/ok to test 30b84a2

@kamran-nvidia
Copy link
Author

/ok to test e927773

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.

3 participants