[Model] Pass param prefix to LLMHead #24862
Merged
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: whx-sjtu <[email protected]>
whx-sjtu
requested review from
patrickvonplaten and
sighingnow
as code owners
mergify
bot
added
deepseek
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request systematically addresses a potential bug related to quantization by ensuring the prefix parameter is passed to ParallelLMHead across all relevant models. The changes are consistent, straightforward, and correctly implemented. This is a valuable fix that enhances the robustness and correctness of the model quantization process. The changes look good.
Signed-off-by: whx-sjtu <[email protected]>
whx-sjtu
force-pushed
the
add_prefix_to_llmhead
branch
from
3c216af to
80aea6b
Compare
|
the prefix is very useful in custom quantization case. |
github-project-automation
bot
moved this from To Triage
to Ready
in gpt-oss Issues & Enhancements
github-project-automation
bot
moved this from Ready
to Done
in gpt-oss Issues & Enhancements
FeiDaLI
pushed a commit
to FeiDaLI/vllm
that referenced
this pull request
Sep 25, 2025
Signed-off-by: whx-sjtu <[email protected]>
charlifu
pushed a commit
to ROCm/vllm
that referenced
this pull request
Sep 25, 2025
Signed-off-by: whx-sjtu <[email protected]> Signed-off-by: charlifu <[email protected]>
5 tasks
xuebwang-amd
pushed a commit
to xuebwang-amd/vllm
that referenced
this pull request
Oct 10, 2025
Signed-off-by: whx-sjtu <[email protected]> Signed-off-by: xuebwang-amd <[email protected]>
choprahetarth
pushed a commit
to Tandemn-Labs/vllm
that referenced
this pull request
Oct 11, 2025
Signed-off-by: whx-sjtu <[email protected]>
xuebwang-amd
pushed a commit
to xuebwang-amd/vllm
that referenced
this pull request
Oct 24, 2025
Signed-off-by: whx-sjtu <[email protected]> Signed-off-by: xuebwang-amd <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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.
Purpose
prefixis an init parameter ofParallelLMHead, which is later used to get_quant_method by VocabParallelEmbedding. Currently only part of models pass this parameter to initializeParallelLMHead. Missing of this parameter can cause bugs in the process of getting quantization method. This PR completes the passing of theprefixparameter for all models.Test Plan
No need to add new test.
Test Result
all tests should pass
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.