Skip to content

Pull requests: vllm-project/vllm

New pull request New
5,350 Open 31,619 Closed
5,350 Open 31,619 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

[Bugfix][Quantization] Match compressed-tensors MoE target aliases bug Something isn't working quantization
#55814 opened Sep 8, 2026 by Kangwenqiao • Draft
[BugFix] GLM-5: load msmodelslim-native checkpoints and drop MTP draft weights bug Something isn't working glm
#55813 opened Sep 8, 2026 by Liears Loading…
[Metrics] Add admission rejection metrics documentation Improvements or additions to documentation
#55812 opened Sep 8, 2026 by thuongvu Loading…
1
[ROCm][Kimi-K3] Add large-M merged MoE front k3 kimi rocm Related to AMD ROCm
#55811 opened Sep 8, 2026 by jiacao-amd Contributor • Draft
[BugFix] Group GLM-5 hybrid KV cache specs (indexer/state roles) correctly bug Something isn't working glm kv-cache-manager
#55810 opened Sep 8, 2026 by Liears Loading…
[Bench] Save server queue time and TTFT in detailed results documentation Improvements or additions to documentation performance Performance-related issues rust
#55809 opened Sep 8, 2026 by lzzzzzc Loading…
4 tasks done
1
[ROCm][Perf] Remove AITER paged-MQA outputs guard for DeepSeek-V4 deepseek Related to DeepSeek models DSv4 rocm Related to AMD ROCm
#55808 opened Sep 8, 2026 by shen-shanshan Collaborator • Draft
4 tasks
[Docs] Add Qwen3.5 text CausalLM to supported models documentation Improvements or additions to documentation qwen Related to Qwen models
#55807 opened Sep 8, 2026 by imitater-dou • Draft
4 tasks done
2
[BugFix] Flatten tuple-valued layer caches in copy_kv_cache_blocks_inplace bug Something isn't working
#55806 opened Sep 8, 2026 by Liears Loading…
1
[Docs] Add DiffusionGemma, LongcatFlashNgram, and Step3Text to supported models documentation Improvements or additions to documentation
#55805 opened Sep 8, 2026 by imitater-dou • Draft
2
[Bugfix][EPLB] Use group-local ranks for torch P2P transfers bug Something isn't working
#55804 opened Sep 8, 2026 by LostFox11 Loading…
1
[Docs] Add FunAudioChatForConditionalGeneration to supported models documentation Improvements or additions to documentation
#55803 opened Sep 8, 2026 by imitater-dou • Draft
2
[Attention] Remove DCP indexer interleave guard and test TP1 output parity glm nvidia
#55802 opened Sep 8, 2026 by stu-cao • Draft
3 tasks done
1
[Bugfix][KV Connector] Apply KV recompute threshold to NIXL remote-prefill pulls bug Something isn't working documentation Improvements or additions to documentation kv-connector
#55801 opened Sep 8, 2026 by LPMing Loading…
2
[Bugfix][CI] skip conftest for NPU compatibility test bug Something isn't working ci/build
#55799 opened Sep 8, 2026 by yzeyu71 Contributor Loading…
1 of 6 tasks
[LoRA] Support PEFT trainable token rows alongside LoRA adapters documentation Improvements or additions to documentation
#55796 opened Sep 8, 2026 by zupengwang Contributor Loading…
2
[Docs] Add LagunaForCausalLM to supported models documentation Improvements or additions to documentation
#55795 opened Sep 8, 2026 by imitater-dou • Draft
4 tasks done
2
[Docs] Add NemotronH_Nano_VL_V2 to supported models documentation Improvements or additions to documentation
#55794 opened Sep 8, 2026 by imitater-dou • Draft
3 tasks
2
[Docs] Add Inkling to supported models documentation Improvements or additions to documentation inkling
#55793 opened Sep 8, 2026 by imitater-dou • Draft
4 tasks
2
[Docs] Add NemotronParse to supported models documentation Improvements or additions to documentation
#55791 opened Sep 8, 2026 by imitater-dou • Draft
3 tasks
2
[Bugfix][Frontend] Clean up child tasks when request wrappers are cancelled bug Something isn't working frontend
#55790 opened Sep 8, 2026 by git-jxj Loading…
1
Revert "[Bugfix][Gemma] Conditionally create KV projections/norms on KV-shared layers" (#54917) bug Something isn't working
#55789 opened Sep 8, 2026 by vllm-agent Contributor • Draft
[Bugfix][Tool Parser] Stream all parallel hy_v4 tool calls arriving in one delta bug Something isn't working tool-calling
#55788 opened Sep 8, 2026 by nikhilkulkarni1755 Contributor Loading…
[Bugfix] FlashInfer derive CUDA-graph head count from KV group layers bug Something isn't working ci/build documentation Improvements or additions to documentation frontend kv-cache-manager kv-connector needs-rebase new-model Requests to new models nvidia performance Performance-related issues qwen Related to Qwen models scheduler speculative-decoding tool-calling
#55787 opened Sep 8, 2026 by XFDG Loading…
1
3
[Bugfix][Frontend] Apply the model's transcription post-processing on the realtime path bug Something isn't working frontend
#55786 opened Sep 8, 2026 by twu3202 Loading…
3 of 4 tasks
1
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.