Fix caching allocator warmup byte estimation for EP model loading#46149
Merged
Conversation
Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
Contributor
Author
|
find in deepseek v4 flash ep run |
Contributor
|
View the CircleCI Test Summary for this PR: https://huggingface.co/spaces/transformers-community/circle-ci-viz?pr=46149&sha=4b2923 |
Contributor
Author
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
kaixuanliu
pushed a commit
to kaixuanliu/transformers
that referenced
this pull request
May 28, 2026
…ggingface#46149) Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
yuchenxie4645
pushed a commit
to yuchenxie4645/transformers
that referenced
this pull request
May 28, 2026
…ggingface#46149) Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
kashif
pushed a commit
to kashif/transformers
that referenced
this pull request
Jun 1, 2026
…ggingface#46149) Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
@ArthurZucker @Cyrilvallez
In EP runs, the effective distributed plan is exposed through model.tp_plan, which switches to the EP plan when distributed_config.enable_expert_parallel is set. Because warmup bypassed that property and read model._tp_plan directly, it could overestimate local device memory and try to preallocate as if expert weights were not sharded.