[BugFix] Continue decode if don't need transfer kv cache between two …#2502
Merged
hsliuustc0106 merged 3 commits intoApr 7, 2026
Merged
Conversation
…stages Signed-off-by: princepride <wangzhipeng628@gmail.com>
Collaborator
Author
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: f6ec96f30a
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
Signed-off-by: princepride <wangzhipeng628@gmail.com>
…tion deserialize_additional_information() reconstructs all entries (including tensors from bytes) on every call. Since _request_omits_kv_transfer_to_next_stage is invoked on each scheduler tick, this caused unnecessary CPU copies and memory churn during decode. Cache the boolean per request and clean up in _free_request. Signed-off-by: princepride <wangzhipeng628@gmail.com> Made-with: Cursor
Collaborator
hsliuustc0106
left a comment
There was a problem hiding this comment.
Has regression tests (test_bagel_understanding.py). Logic looks correct.
Collaborator
Author
|
@hsliuustc0106 Can you help approve it? |
skf-1999
pushed a commit
to Semmer2/vllm-omni
that referenced
this pull request
Apr 7, 2026
vllm-project#2502) Signed-off-by: princepride <wangzhipeng628@gmail.com>
vraiti
pushed a commit
to vraiti/vllm-omni
that referenced
this pull request
Apr 9, 2026
vllm-project#2502) Signed-off-by: princepride <wangzhipeng628@gmail.com>
bob-021206
pushed a commit
to jasonlee-1024/vllm-omni
that referenced
this pull request
Apr 21, 2026
vllm-project#2502) Signed-off-by: princepride <wangzhipeng628@gmail.com> Signed-off-by: bob-021206 <binyan_github@163.com>
lengrongfu
pushed a commit
to lengrongfu/vllm-omni
that referenced
this pull request
May 1, 2026
vllm-project#2502) Signed-off-by: princepride <wangzhipeng628@gmail.com>
clodaghwalsh17
pushed a commit
to clodaghwalsh17/nm-vllm-omni-ent
that referenced
this pull request
May 12, 2026
vllm-project#2502) Signed-off-by: princepride <wangzhipeng628@gmail.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
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.
…stages
PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED.
Purpose
While the previous update resolved the synchronization error in KV cache transfers across stages, it introduced a side effect where Stage 0 output is terminated unconditionally, even when no transmission is occurring and
img2textandtext2texttask can not output any text.Test Plan
Test Result
** before: **
** after: **
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model. Please runmkdocs serveto sync the documentation editions to./docs.BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)