Skip to content

Conversation

@malay-nagda
Copy link
Contributor

@malay-nagda malay-nagda commented Jan 22, 2026

What does this PR do ?

Disable precision aware optimizer when used with FSDP

Changelog

if compute_dtype == "bf16" and not recipe.ddp.use_megatron_fsdp:
        recipe.optimizer.use_precision_aware_optimizer = True

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)

@malay-nagda malay-nagda self-assigned this Jan 22, 2026
@malay-nagda malay-nagda added bug Something isn't working performance performance/release Performance items related with NeMo release labels Jan 22, 2026
@malay-nagda malay-nagda added this to the 26.02 milestone Jan 22, 2026
@copy-pr-bot
Copy link

copy-pr-bot bot commented Jan 22, 2026

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

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

Labels

bug Something isn't working performance/release Performance items related with NeMo release performance

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants