update: change kv cache to meta tensor until device tensor is fully supported#306
Merged
rebel-jonghewk merged 1 commit intodev-0.12from Jan 30, 2026
Merged
update: change kv cache to meta tensor until device tensor is fully supported#306rebel-jonghewk merged 1 commit intodev-0.12from
rebel-jonghewk merged 1 commit intodev-0.12from
Conversation
rebel-jonghewk
requested review from
rebel-eunji,
rebel-jaehwang,
rebel-jiwoopark,
rebel-jongho,
rebel-myeongbo and
rebel-wonsubkim
rebel-jonghewk
changed the title
rebel-jongho
approved these changes
Jan 30, 2026
rebel-jaehwang
pushed a commit
that referenced
this pull request
Jan 30, 2026
rebel-jaehwang
pushed a commit
that referenced
this pull request
Jan 30, 2026
rebel-jaehwang
pushed a commit
that referenced
this pull request
Jan 30, 2026
rebel-wonsubkim
added a commit
that referenced
this pull request
Feb 2, 2026
This reverts commit af04acc.
rebel-wonsubkim
added a commit
that referenced
this pull request
Feb 2, 2026
rebel-jaehwang
pushed a commit
that referenced
this pull request
Feb 2, 2026
rebel-jiwoopark
pushed a commit
that referenced
this pull request
Feb 4, 2026
rebel-jiwoopark
pushed a commit
that referenced
this pull request
Feb 4, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
🚀 Summary of Changes
While device tensor (rbln) support in vLLM is under development, this PR replaces the CPU-based KV cache with meta-device tensors to prevent unnecessary host memory usage.
This will require latest rebel-compiler to work.
📌 Related Issues / Tickets
✅ Type of Change
feature)model)core)bug-fix)perf)refactor)docs)other): please describe🧪 How to Test
.........📸 Screenshots / Logs (if applicable)
📋 Checklist
💬 Notes