Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,263 Open 15,899 Closed
1,263 Open 15,899 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] Add offline FastAPI documentation support for air-gapped environments frontend
#30184 opened Dec 6, 2025 by rickychen-infinirc Loading…
1
5
[WIP][Feat][Sched] Add Buffered_Response v1
#30183 opened Dec 6, 2025 by Pr0Wh1teGivee Loading…
5 tasks
9
[MISC]: change NIXL compatibility hash logging level to debug kv-connector
#30182 opened Dec 6, 2025 by AuruTus Loading…
[Model] Move multimodal_cpu_fields definition to field config multi-modality Related to multi-modality (#4194) qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed tpu Related to Google TPUs v1
#30181 opened Dec 6, 2025 by DarkLight1337 Loading…
5 tasks
6
[Bugfix] fix fuse_allreduce_rms when tp =1
#30178 opened Dec 6, 2025 by ZJY0516 Loading…
5 tasks
2
[ROCm][MXFP4] Enable FP4 MLA BMM support rocm Related to AMD ROCm v1
#30177 opened Dec 6, 2025 by dllehr-amd Loading…
5 tasks
8
[Misc][Core] Remove unused req_index increment in scheduler v1
#30176 opened Dec 6, 2025 by ivanium Loading…
5 tasks
[Frontend] Add --uvicorn-access-log-exclude-paths option frontend
#30175 opened Dec 6, 2025 by GeoffreyWang1117 Loading…
4 tasks done
1
2
[Bugfix] Improve DCP error message with backend hint v1
#30174 opened Dec 6, 2025 by GeoffreyWang1117 Loading…
2 tasks done
1
3
[BugFix] Fix assert batch_descriptor.num_tokens == num_tokens_padded nvidia ready ONLY add when PR is ready to merge/full CI is needed v1
#30173 opened Dec 6, 2025 by LucasWilkinson Loading…
4
[Frontend] Remove confusing -O.xx flag error ready ONLY add when PR is ready to merge/full CI is needed
#30169 opened Dec 6, 2025 by gmagogsfm Loading…
1
[Core][Hybrid allocator + connector] Support hybrid allocator + kv cache connector kv-connector tpu Related to Google TPUs v1
#30166 opened Dec 6, 2025 by ivanium Loading…
5 tasks
7
Nvidia ModelOpt workaround for issue 28072 nvidia quantization
#30164 opened Dec 6, 2025 by shengliangxu Loading…
@shengliangxu
7
[Deepseek] Fix OOM during DeepSeek R1 startup deepseek Related to DeepSeek models v1
#30162 opened Dec 5, 2025 by MatthewBonanni Loading…
3 of 5 tasks
5
[CI] Update Test Dependencies ci/build
#30160 opened Dec 5, 2025 by junpuf Loading…
5 tasks
[Perf] Optimize group_topk kernel, 1.9% Throughput improvement, 2.1% TPOT improvemnt ready ONLY add when PR is ready to merge/full CI is needed
#30159 opened Dec 5, 2025 by yewentao256 Loading…
7
[responsesAPI][6] input/output messages for ResponsesParser frontend gpt-oss Related to GPT-OSS models
#30158 opened Dec 5, 2025 by qandrew • Draft
5 tasks
6
feat: add TxtSlicesDataset to allow sampling slices from txt file for benchmarking performance Performance-related issues
#30156 opened Dec 5, 2025 by hypdeb Loading…
8
update torchao safetensors impl ready ONLY add when PR is ready to merge/full CI is needed
#30155 opened Dec 5, 2025 by liangel-02 Loading…
4
Integration for Ray LLM with load_format=runai_streamer
#30154 opened Dec 5, 2025 by jiangwu300 Loading…
5 tasks
6
Bump nvshmem to 3.3.24 and fix CUDA 13 installation nvidia
#30149 opened Dec 5, 2025 by dmitry-tokarev-nv Loading…
5 tasks
3
[WIP] Nightly signal on torch ci/build nvidia rocm Related to AMD ROCm
#30146 opened Dec 5, 2025 by atalman Loading…
5 tasks
6
[Renderer] Separate out RendererConfig from ModelConfig deepseek Related to DeepSeek models documentation Improvements or additions to documentation frontend kv-connector llama Related to Llama models multi-modality Related to multi-modality (#4194) qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed ready-run-all-tests Trigger CI with all tests for wide-ranging PRs speculative-decoding structured-output tpu Related to Google TPUs v1
#30145 opened Dec 5, 2025 by DarkLight1337 Loading…
5 tasks
44
[Misc] Remove pad_for_cudagraphs from config needs-rebase nvidia speculative-decoding v1
#30143 opened Dec 5, 2025 by LucasWilkinson Loading…
6
Add llmcompressor fp8 kv-cache quant (per-tensor and per-attn_head) documentation Improvements or additions to documentation llama Related to Llama models needs-rebase speculative-decoding v1
#30141 opened Dec 5, 2025 by eldarkurtic Loading…
13
ProTip! no:milestone will show everything without a milestone.