docs: qualify SALM Automodel parallelism support - #15976
Merged
Merged
Conversation
Signed-off-by: Dongji Gao <dongjig@nvidia.com>
DongjiGao
requested review from
KunalDhawan,
blisc and
pzelasko
and removed request for
pzelasko
July 28, 2026 18:51
DongjiGao
marked this pull request as ready for review
July 28, 2026 18:55
blisc
added
the
r3.0.0
Collaborator
|
Do you want to merge or keep as draft? |
Contributor
Author
I think this one is good to merge. |
blisc
pushed a commit
that referenced
this pull request
Jul 28, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changed
AutomodelParallelStrategycan construct meshes for FSDP2, TP, PP, CP, EP, and HSDP, while actual support remains model-, backend-, and combination-dependent.tp_size,dp_replicate_size, and the currently unvalidatedpp_sizeat 1 for this configuration.Why
The previous blanket support statement drove QA and users toward unsupported Nemotron-V3 configurations. Nemotron-V3 has no TP plan, while the SALMAutomodel HSDP path is unsupported and currently fails with an opaque DeviceMesh error. PP is capability-signaled by Automodel but has no SpeechLM2 training validation.
This is a documentation correction only; it does not change parallelism behavior. The HSDP validation/runtime issue remains separate.
Validation
sphinx-lint docs/source/speechlm2/training_and_scaling.rst— no problems foundgit diff --checkRelated: internal bug 6437904.