Skip to content

Pull requests: vllm-project/vllm

New pull request New
2,074 Open 20,803 Closed
2,074 Open 20,803 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

[Feature]: IndexCache support for DSA models deepseek Related to DeepSeek models
#37735 opened Mar 21, 2026 by chaunceyjiang • Draft
5 tasks
1
2
Support FP8 KVCache on XPU v1
#37731 opened Mar 21, 2026 by xinyu-intel • Draft
5 tasks
2
Fix Mamba state corruption from stale CUDA graph block table entries fb-exported meta-exported nvidia v1
#37728 opened Mar 21, 2026 by minosfuture Loading…
3
[Bugfix] Fix Responses API instructions leaking through previous_response_id bug Something isn't working frontend
#37727 opened Mar 21, 2026 by he-yufeng Loading…
1
2
Revert "[Model] Deprecate the score task (this will not affect users)." (#37537) documentation Improvements or additions to documentation frontend v1
#37726 opened Mar 21, 2026 by zhewenl • Draft
5
[Bugfix] Preserve CUDA arch suffix (a/f) for SM12x — fixes NVFP4 NaN on desktop Blackwell bug Something isn't working ci/build nvidia ready ONLY add when PR is ready to merge/full CI is needed
#37725 opened Mar 20, 2026 by RobTand Loading…
2 tasks done
7
[ROCm][CI] Stabilize ROCm speech-to-text translation test with ROCM_EXTRA_ARGS ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
#37723 opened Mar 20, 2026 by AndreasKaratzas • Draft
1
quick fix for 37665
#37722 opened Mar 20, 2026 by xuechendi Loading…
5 tasks
[ROCm][CI] Update GSM8K eval config to use fp8-and-mixed models list (MI355) ci/build ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
#37721 opened Mar 20, 2026 by AndreasKaratzas Loading…
1
[Test] Only Run MLA model when user explicitly set for batch invariance ready ONLY add when PR is ready to merge/full CI is needed v1
#37719 opened Mar 20, 2026 by yewentao256 Loading…
2
[Bug] Fix fp8 deepgemm batch invariant bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed
#37718 opened Mar 20, 2026 by yewentao256 Loading…
[ROCm][CI] Add large_gpu_mark to test_max_tokens_none for ROCm ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
#37717 opened Mar 20, 2026 by AndreasKaratzas Loading…
2
[BugFix] Fix MoRIIOConnector for disaggregated P/D inference bug Something isn't working kv-connector
#37716 opened Mar 20, 2026 by raviguptaamd • Draft
3
[Bugfix] Mask padded rows for FlashInfer CUTLASS MoE bug Something isn't working nvidia
#37715 opened Mar 20, 2026 by kjiang249 • Draft
5 tasks
2
Readability cleanup for wvSplitK reduces. rocm Related to AMD ROCm
#37713 opened Mar 20, 2026 by amd-hhashemi Loading…
5 tasks
Properly enable wvSplitK fp8 path for RDNA
#37712 opened Mar 20, 2026 by amd-hhashemi Loading…
5 tasks
1
[Bugfix] Fix structured output crash on CPU due to pin_memory=True bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed structured-output v1
#37706 opened Mar 20, 2026 by wjhrdy Loading…
3 tasks done
1
6
[Test] Add more unittests for CUDAGraphWrapper nvidia v1
#37702 opened Mar 20, 2026 by SoluMilken Loading…
3 of 5 tasks
1
[Bugfix] Fix FLA Hopper/TMA misclassification on SM12x desktop Blackwell bug Something isn't working
#37700 opened Mar 20, 2026 by RobTand Loading…
3 tasks done
1
5
[Bugfix] Respect VLLM_WEIGHT_OFFLOADING_DISABLE_PIN_MEMORY in prefetch offloader bug Something isn't working
#37699 opened Mar 20, 2026 by he-yufeng Loading…
1
9
[ROCm][Bugfix] fix exception related to trust_remote_code for MiniMax-M2.1-MXFP4 bug Something isn't working cpu Related to CPU backends rocm Related to AMD ROCm
#37698 opened Mar 20, 2026 by hongxiayang Loading…
5 tasks
4
[torch.compile]: Disable Sequence Parallelism (SP) for piecewise compilation v1
#37696 opened Mar 20, 2026 by SouthWest7 Loading…
3 of 5 tasks
5
[Perf] Use torch compile to fuse pack topk in trtllm moe nvidia performance Performance-related issues ready ONLY add when PR is ready to merge/full CI is needed
#37695 opened Mar 20, 2026 by wzhao18 Loading…
5 tasks
6
[FlexAttention] allow custom mask mod v1
#37692 opened Mar 20, 2026 by liangel-02 Loading…
3
[cpu][ci] remove soft-fail for Arm CI and add quant model tests ci/build cpu Related to CPU backends
#37691 opened Mar 20, 2026 by fadara01 Loading…
2 tasks
@fadara01
4
ProTip! Follow long discussions with comments:>50.