fix: load HF model only on rank 0 #544
Open
+52
−5
Conversation
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
Signed-off-by: Parth Chadha <pchadha@nvidia.com>
terrykong
reviewed
Jun 24, 2025
aae142f to
ebaeb99
Compare
terrykong
reviewed
Jun 24, 2025
Signed-off-by: Parth Chadha <pchadha@nvidia.com>
ebaeb99 to
fcec2db
Compare
terrykong
previously approved these changes
Jun 24, 2025
Any commits made after this event will not be merged.
Any commits made after this event will not be merged.
Signed-off-by: Parth Chadha <pchadha@nvidia.com>
Signed-off-by: Parth Chadha <pchadha@nvidia.com>
terrykong
approved these changes
Jun 27, 2025
Any commits made after this event will not be merged.
auto-merge was automatically disabled
June 28, 2025 02:16
Pull Request is not mergeable
Any commits made after this event will not be merged.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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 does this PR do ?
Loads model only on rank 0 and then uses fsdp2
set_model_state_dictAPI to load the weights on other ranks (after the model has been parallelized).Note that the current PR still leads to GPU OOM for 70B model on 1 node with dtensor, will fix in a separate PR.
Issues
List issues that this PR closes (syntax):
Usage
# Add a code snippet demonstrating how to use thisBefore your PR is "Ready for review"
Pre checks:
Additional Information