Skip to content

Pull requests: vllm-project/vllm

New pull request New
2,488 Open 15,503 Closed
2,488 Open 15,503 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

Gemma4 HeteroTP support for NIXL KV connector kv-connector needs-rebase v1
#41169 opened Apr 28, 2026 by ZhanqiuHu Contributor • Draft
1
[Bugfix] Fix DeepSeek V4 MTP=1 hang issue bug Something isn't working deepseek Related to DeepSeek models v1
#41168 opened Apr 28, 2026 by wzhao18 Contributor • Draft
4 tasks
[security] Change VLLM_MEDIA_URL_ALLOW_REDIRECTS default to False
#41167 opened Apr 28, 2026 by Dadaam Loading…
6
[Ci][BugFix] Fix slow DP tests due to bad teardown logic bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed v1
#41166 opened Apr 28, 2026 by njhill Member Loading…
3
[ROCm][Bugfix][GPTOSS]: fix input_ids and expert_map args for quark w4a8 gptoss bug Something isn't working gpt-oss Related to GPT-OSS models rocm Related to AMD ROCm
#41165 opened Apr 28, 2026 by Rohan138 Contributor Loading…
4 tasks
3
[DP] External DP LB compatibility with all GPU visibility documentation Improvements or additions to documentation v1
#41164 opened Apr 28, 2026 by njhill Member Loading…
4
[Perf] Optimize AllPool.forward by slicing first, 51% faster in the method level benchmark ready ONLY add when PR is ready to merge/full CI is needed
#41163 opened Apr 28, 2026 by yewentao256 Member Loading…
3
[Model Runner V2] Rebuild attn metadata between draft decode steps v1
#41162 opened Apr 28, 2026 by TheEpicDolphin Collaborator • Draft
3
Fix static actorder handling for compressed-tensors WNA16 MoE
#41161 opened Apr 28, 2026 by ZewenShen-Cohere Loading…
6 tasks done
7
[Kernel] Pack output and LSE in DCP A2A ready ONLY add when PR is ready to merge/full CI is needed v1
#41160 opened Apr 28, 2026 by sungsooha Contributor Loading…
4 tasks done
6
[ROCm] AITER config object rocm Related to AMD ROCm
#41159 opened Apr 28, 2026 by gshtras Collaborator • Draft
2
[UX] Enable PyTorch expandable_segments by default v1
#41158 opened Apr 28, 2026 by MatthewBonanni Collaborator • Draft
4 tasks
2
[Model] Add Apertus Tool Parser documentation Improvements or additions to documentation tool-calling
#41154 opened Apr 28, 2026 by blancsw Loading…
3 of 5 tasks
6
[Perf][DeepSeek V4 MTP] Run e_proj and h_proj in parallel streams deepseek Related to DeepSeek models
#41151 opened Apr 28, 2026 by zixi-qi Collaborator • Draft
2 tasks
4
[Model][DeepSeekV4] Use FlashMLA KV-cache path for prefill deepseek Related to DeepSeek models ready ONLY add when PR is ready to merge/full CI is needed v1
#41150 opened Apr 28, 2026 by PerkzZheng Loading…
4
[CI/Build] Auto-detect manylinux ABI tag for nightly wheels ci/build ready ONLY add when PR is ready to merge/full CI is needed
#41149 opened Apr 28, 2026 by Harry-Chen Member Loading…
3 of 4 tasks
18
[Bugfix] Fix repeated DSv4 RoPE cache initialization bug Something isn't working deepseek Related to DeepSeek models
#41148 opened Apr 28, 2026 by jeejeelee Collaborator Loading…
4 tasks
3
[CI] De-flake test_chat_completion_n_parameter_non_streaming
#41147 opened Apr 28, 2026 by haosdent Contributor Loading…
2
better logging for large uncachable items multi-modality Related to multi-modality (#4194)
#41145 opened Apr 28, 2026 by h-avsha Contributor Loading…
3
[Bugfix][Qwen3-Next] Force a/b contiguity in fix_query_key_value_ordering bug Something isn't working qwen Related to Qwen models
#41142 opened Apr 28, 2026 by Yeuvoir Loading…
3 tasks
1
1
[EPLB] Add offline mapping support documentation Improvements or additions to documentation
#41141 opened Apr 28, 2026 by arpera Contributor Loading…
2 of 4 tasks
1
First draft of LLM assisted test target determination ci/build
#41139 opened Apr 28, 2026 by avinashsingh77 Contributor Loading…
4 tasks
4
[XPU][Mamba] Add Triton-based selective scan forward op for XPU intel-gpu Related to Intel GPU
#41137 opened Apr 28, 2026 by mfylcek Contributor • Draft
4 tasks
4
[ROCm] DeepSeekV4-Flash-Base model enablement on ROCm with triton & torchfallback deepseek Related to DeepSeek models kv-connector rocm Related to AMD ROCm v1
#41136 opened Apr 28, 2026 by lcskrishna Contributor Loading…
4 tasks
2
[Bugfix] fix inductor error for dpsk v4 bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed v1
#41135 opened Apr 28, 2026 by ZJY0516 Member Loading…
4 tasks
1
2
ProTip! Updated in the last three days: updated:>2026-04-25.