NVIDIA / NVIDIA/Megatron-LM

Integrate MFSDP v2 with Megatron Bridge

Open
#6,325 0 comments 0 reactions 1 assignee Claimed by @wujingyue View on GitHub
Dominant language
Python
Stars
17.9k
Forks
4.5k
Avg merge
4d 6h
Merged PRs (30d)
271

Description

## Goal

Enable Megatron Bridge training recipes to use the experimental MFSDP v2 path.

## Scope

Build on the Megatron Core integration from #5614 and add the Bridge-side configuration and recipe plumbing needed to select MFSDP v2.

## Intended outcome

- Expose MFSDP v2 through an experimental Bridge configuration.
- Map the Bridge configuration to the MFSDP v2/MCore training path.
- Preserve existing FSDP behavior when MFSDP v2 is not selected.
- Add a minimal Bridge smoke test or recipe that verifies training can start with MFSDP v2 enabled.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.