Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,136 Open 12,999 Closed
1,136 Open 12,999 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

[Core] Bookkeeping optimization: Vectorize updates v1
#25801 opened Sep 27, 2025 by Jialin Loading…
3 of 5 tasks
[Feat][EPLB][Perf] Enable Round-robin expert placement strategy while eplb is enabled.
#25798 opened Sep 27, 2025 by cboss6 Loading…
2
[Bugfix] Add missing image_size for phi4_multimodal ready ONLY add when PR is ready to merge/full CI is needed
#25796 opened Sep 27, 2025 by Renovamen Loading…
5
[Misc] Update openai client example file for multimodal documentation Improvements or additions to documentation ready ONLY add when PR is ready to merge/full CI is needed
#25795 opened Sep 27, 2025 by ywang96 Loading…
5 tasks
3
Add filtering for chat template kwargs frontend ready ONLY add when PR is ready to merge/full CI is needed
#25794 opened Sep 27, 2025 by russellb Loading… v0.11.0 Cherry Picks
3
[DRAFT][DO NOT REVIEW] Compile once per block qwen Related to Qwen models
#25792 opened Sep 26, 2025 by Lucaskabela • Draft
5 tasks
perf: optimize rejection sampling triton kernel v1
#25791 opened Sep 26, 2025 by happierpig Loading…
5 tasks
2
[gpt-oss] disable tool server initialization if no tool in request frontend gpt-oss Related to GPT-OSS models
#25790 opened Sep 26, 2025 by qandrew Loading…
5
[Bugfix] Fix hang with DP+EP on B200 v1
#25789 opened Sep 26, 2025 by alexm-redhat Loading…
2
[Benchmark] Cleanup deprecated nightly benchmark and adjust the docstring for performance benchmark ci/build documentation Improvements or additions to documentation performance Performance-related issues
#25786 opened Sep 26, 2025 by KuntaiDu Loading…
5 tasks
9
[Misc] Make EP kernels install script support uv
#25785 opened Sep 26, 2025 by LucasWilkinson Loading…
4
[Misc] Integrate Suffix Decoding from Arctic Inference needs-rebase v1
#25784 opened Sep 26, 2025 by aurickq Loading…
1 of 3 tasks
8
Validate API tokens in constant time frontend ready ONLY add when PR is ready to merge/full CI is needed
#25781 opened Sep 26, 2025 by russellb Loading… v0.11.0 Cherry Picks
3
Hybrid deepep
#25778 opened Sep 26, 2025 by bnellnm • Draft
5 tasks
[BugFix] Potential B200 + full-CG hang fix
#25776 opened Sep 26, 2025 by LucasWilkinson • Draft
5 tasks
[Bugfix] Improve GPU validation logging in Ray fallback scenarios
#25775 opened Sep 26, 2025 by sairampillai Loading…
3 of 5 tasks
1
2
fuse rope ci/build tpu Related to Google TPUs v1
#25774 opened Sep 26, 2025 by PatrykSaffer • Draft
Fix GPTQ model loading in Transformers backend
#25770 opened Sep 26, 2025 by hmellor Loading…
2
Add batch invariant kernel override for FlashInfer backend v1
#25769 opened Sep 26, 2025 by bwasti Loading…
3 of 5 tasks
4
DP Coordination refactor v1
#25768 opened Sep 26, 2025 by SageMoore Loading…
[Bug]: Set LD_LIBRARY_PATH to include the 'standard' CUDA location ci/build
#25766 opened Sep 26, 2025 by smarterclayton Loading…
5 tasks
1
3
[CI] Push multiarch manifests as nightly builds ci/build
#25764 opened Sep 26, 2025 by csahithi Loading…
5 tasks
7
[ROCm][Perf] New design on ROCm AITER MHA backend Implementation rocm Related to AMD ROCm v1
#25763 opened Sep 26, 2025 by ganyi1996ppo Loading…
5 tasks
9
[Core] Fix torch.dynamo compatibility for Qwen models on vllm-gaudi qwen Related to Qwen models v1
#25761 opened Sep 26, 2025 by pawel-olejniczak • Draft
5 tasks done
1
Refactor/understanding prepare inputs padded speculative-decoding v1
#25758 opened Sep 26, 2025 by tomasruizt • Draft
ProTip! Filter pull requests by the default branch with base:main.