Skip to content

Fix vllm weight mapping import issue. #2629

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix vllm weight mapping import issue. #2629

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions src/MaxText/integration/tunix/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,11 +16,11 @@

import re

import maxtext.src.maxtext.integration.tunix.weight_mapping as weight_mapping # pylint: disable=consider-using-from-import
from MaxText.integration.tunix.weight_mapping import StandaloneVllmWeightMapping
Copy link
Collaborator

@bvandermoon bvandermoon Nov 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you know why this happened with copybara?

Copy link
Collaborator

@A9isha A9isha Nov 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@abhinavclemson you would need to update the copybara script otherwise this PR will not be useful

Also there are other places where this is broken, so other files need to be added to the copybara
https://github.com/AI-Hypercomputer/maxtext/compare/anisha-fix-trainrl?expand=1

from MaxText.utils.ckpt_conversion.utils.param_mapping import PARAM_MAPPING
from MaxText.utils.ckpt_conversion.utils.param_mapping import VLLM_HOOK_FNS

STANDALONE_VLLM_WEIGHT_MAPPING = weight_mapping.StandaloneVllmWeightMapping()
STANDALONE_VLLM_WEIGHT_MAPPING = StandaloneVllmWeightMapping()

# This static map provides the architectural knowledge (sharding) that is
# not present in the original HF mapping.
Expand Down
Loading