modelopt=0.21.0 update

[janekl]

What does this PR do ?

• Bugfix to correctly support Mixtral for PTQ in NeMo 2.0. done in Pass the number of experts to modelopt layer spec #11607
• Updating nvidia-modelopt=0.21.0

Collection: LLM / NLP

Changelog

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

Usage

• You can potentially add a usage example below

# Add a code snippet demonstrating how to use this 

GitHub Actions CI

The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.

The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".

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?

PR Type:

• New Feature
• Bugfix
• Documentation

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

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

• Related to Pass the number of experts to modelopt layer spec #11607

[github-actions]

[🤖]: Hi @janekl 👋,

We wanted to let you know that a CICD pipeline for this PR just finished successfully

So it might be time to merge this PR or get some approvals

I'm just a bot so I'll leave it you what to do next.

//cc @pablo-garay @ko3n1g

[ko3n1g]

Why is modelopt not part of requirements txt?

[janekl]

Why is modelopt not part of requirements txt?

I think Eric wanted it to be only an optional requirement for NeMo in the initial integration effort (see here and here). It's also import guarded (see here, for example).

Are you suggesting to revisit this decision?

[ko3n1g]

Why is modelopt not part of requirements txt?

I think Eric wanted it to be only an optional requirement for NeMo in the initial integration effort (see here and here). It's also import guarded (see here, for example).

Are you suggesting to revisit this decision?

If this was for initial implementation then yes, let’s revisit it. But not necessarily as part of this PR. Thanks for explaining!

[janekl]

This needs to be synced with an update on container side, I'll let you know once ready

[github-actions]

[🤖]: Hi @janekl 👋,

We wanted to let you know that a CICD pipeline for this PR just finished successfully

So it might be time to merge this PR or get some approvals

I'm just a bot so I'll leave it you what to do next.

//cc @pablo-garay @ko3n1g