Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,280 Open 16,324 Closed
1,280 Open 16,324 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Don'e assume position_embedding_type will be present for BERT and RoBERTa models
#30770 opened Dec 16, 2025 by hmellor Loading…
1
[Frontend] Add max-completion-token option to transcription/translation endpoints frontend
#30769 opened Dec 16, 2025 by NickLucche Loading…
3
Fix instantiation of HfHubHTTPError in LoRA test
#30768 opened Dec 16, 2025 by hmellor Loading…
5
Algo
#30767 opened Dec 16, 2025 by Mercykid-bash Loading…
5 tasks
6
[Doc][CPU] Update CPU doc ci/build documentation Improvements or additions to documentation
#30765 opened Dec 16, 2025 by bigPYJ1151 Loading…
1 of 5 tasks
6
Remove head_mask from Ultravox and Swin ready ONLY add when PR is ready to merge/full CI is needed
#30764 opened Dec 16, 2025 by hmellor Loading…
3
[KVConnector]: Enable Cross-layers KV cache layout for MultiConnector kv-connector
#30761 opened Dec 16, 2025 by kfirtoledo Loading…
5
[WIP]verify_is_multiproc
#30760 opened Dec 16, 2025 by 1643661061leo Loading…
5 tasks
3
[MM] Pass FA version in ViT Attn ready ONLY add when PR is ready to merge/full CI is needed
#30756 opened Dec 16, 2025 by NickLucche Loading…
3
Fix overflow in topk.
#30754 opened Dec 16, 2025 by dcampora Loading…
5 tasks
4
[Core][KVConnector] Propagate block hashes in SchedulerOutput tpu Related to Google TPUs v1
#30753 opened Dec 16, 2025 by QierLi Loading…
3 of 5 tasks
1
[Bugfix]: prevent leaking tokens in crash log v1
#30751 opened Dec 16, 2025 by dr75 Loading…
5 tasks
3
[Refactor] [4/N] Move VLLM_SERVER_DEV endpoints into the serve directory ci/build frontend
#30749 opened Dec 16, 2025 by chaunceyjiang Loading…
5 tasks
10
[Docs] fix function name documentation Improvements or additions to documentation
#30748 opened Dec 16, 2025 by lengrongfu Loading…
5 tasks
4
[Fix] uniform decode batch check v1
#30747 opened Dec 16, 2025 by Jialin Loading…
3 of 5 tasks
3
[Draft][SM100] Enable fp8 compute for prefill MLA v1
#30746 opened Dec 16, 2025 by pavanimajety • Draft
2 of 3 tasks
5
[BugFix]Reclaim resources to prevent memory leaks when use LMCacheMPConnector kv-connector
#30745 opened Dec 16, 2025 by wz1qqx Loading…
5 tasks
5
[BugFix] Fix memory spike in workspace allocation ci/build ready ONLY add when PR is ready to merge/full CI is needed v1
#30744 opened Dec 16, 2025 by LucasWilkinson Loading…
4
[compile] Recompile graph module during Dynamo cache loading.
#30743 opened Dec 16, 2025 by zhxchen17 • Draft
5 tasks
3
Support LoRA of PLaMo 2/3 documentation Improvements or additions to documentation
#30742 opened Dec 16, 2025 by Alnusjaponica • Draft
5 tasks
1
Properly handle packed_modules_mapping of PLaMo2
#30740 opened Dec 16, 2025 by Alnusjaponica • Draft
1 of 5 tasks
2
[BugFix] Support online dense model DP without overhead kv-connector ready ONLY add when PR is ready to merge/full CI is needed v1
#30739 opened Dec 16, 2025 by njhill Loading…
2
1
Add mfu stats logging ready ONLY add when PR is ready to merge/full CI is needed v1
#30738 opened Dec 16, 2025 by SungMinCho Loading…
9
add Qwen3OmniMoeAudioEncoder and support torch compile qwen Related to Qwen models
#30735 opened Dec 16, 2025 by XiaobingSuper Loading…
2 of 5 tasks
9
[CI/Build] Allow user to configure NVSHMEM version via ENV or command line ci/build
#30732 opened Dec 16, 2025 by eicherseiji Loading…
5
ProTip! Updated in the last three days: updated:>2025-12-13.