Skip to content

Pull requests: vllm-project/vllm

New pull request New
2,529 Open 22,229 Closed
2,529 Open 22,229 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

[Docs] [QeRL] Layerwise Reloading Documentation documentation Improvements or additions to documentation
#40317 opened Apr 20, 2026 by kylesayrs Contributor Loading…
1
[Docs] Fix thinking_token_budget docs documentation Improvements or additions to documentation
#40316 opened Apr 20, 2026 by milesial Contributor Loading…
1
fix: Do not make function calls when request has no tools for /v1/responses frontend ready ONLY add when PR is ready to merge/full CI is needed
#40314 opened Apr 20, 2026 by terrytangyuan Contributor Loading…
Revert "Fix MoE backend selection for LoRA (unquantized MoE)" (#40273)
#40313 opened Apr 20, 2026 by vllm-agent • Draft
1
[Model] Use AutoWeightsLoader for GPT2
#40312 opened Apr 20, 2026 by cben484 Loading…
4
[Bugfix] Align chat completion tool_choice/tools validation with OpenAI spec bug Something isn't working frontend tool-calling
#40311 opened Apr 20, 2026 by sfeng33 Collaborator Loading…
3
[Bugfix] Fix W4A8_FP8 MoE tp>1 correctness and view() TypeError bug Something isn't working
#40310 opened Apr 20, 2026 by EdalatiAli Contributor • Draft
4 tasks
3
[QeRL] Add warnings for extra memory buffering
#40309 opened Apr 20, 2026 by kylesayrs Contributor Loading…
5
[Bugfix] Fix hybrid KV manager for quantized per-token-head KV cache bug Something isn't working v1
#40308 opened Apr 20, 2026 by lesj0610 Loading…
1
[ROCm][CI] Fix trust_remote_code AttributeError in EAGLE3 acceptance length test rocm Related to AMD ROCm speculative-decoding v1
#40306 opened Apr 19, 2026 by AndreasKaratzas Collaborator Loading…
1 task done
feat: mla prefill trtllm static fp8 output
#40304 opened Apr 19, 2026 by carlyou Contributor • Draft
4 tasks
4
[Bug] Fix shm_broadcast PyCFunction descriptor corruption under JIT loads bug Something isn't working
#40303 opened Apr 19, 2026 by jsboige Loading…
1 of 4 tasks
1
8
[ROCm][Bugfix] Fall back when Quark MoE AITER dispatch is unsupported bug Something isn't working rocm Related to AMD ROCm
#40300 opened Apr 19, 2026 by Bortlesboat Contributor • Draft
1
4
Register parsed config classes before tokenizer init needs-rebase
#40299 opened Apr 19, 2026 by Bortlesboat Contributor • Draft
1
3
Implement incremental streaming for minimax_m2_tool_parser tool-calling
#40298 opened Apr 19, 2026 by frankie-ys Contributor Loading…
1 of 4 tasks
1
5
[Platform][XPU] Opt-in integrated-GPU override for unified memory intel-gpu Related to Intel GPU
#40295 opened Apr 19, 2026 by MegaStood Loading…
1
[Bugfix] Fix GPTQ not picking up desc_asc from model config bug Something isn't working
#40294 opened Apr 19, 2026 by bajceta Loading…
4 tasks done
4
[ROCm] [Release] Clean-up Release Pipeline ci/build rocm Related to AMD ROCm
#40293 opened Apr 19, 2026 by tjtanaa Collaborator Loading…
4 tasks
3
[ROCm][ViT] Detect Triton-AMD kernels at their new aiter location rocm Related to AMD ROCm
#40289 opened Apr 19, 2026 by Lafunamor Loading…
8
[Bugfix] Fix dataset name and path argument validation bug in vllm bench serve bug Something isn't working performance Performance-related issues verified Run pre-commit for new contributors without triggering other tests
#40288 opened Apr 19, 2026 by talorabr Contributor Loading…
4 tasks
4
[XPU] Update nixl to v0.10.1 in Dockerfile ci/build intel-gpu Related to Intel GPU kv-connector
#40287 opened Apr 19, 2026 by zhenwei-intel Contributor • Draft
4 tasks
5
Fix DP internal LB first-request routing drift v1
#40285 opened Apr 19, 2026 by Abhicodeitout Loading…
3 of 4 tasks
5
Add Granite 4.1 Vision as built-in multimodal model documentation Improvements or additions to documentation multi-modality Related to multi-modality (#4194) new-model Requests to new models
#40282 opened Apr 19, 2026 by artem-spector Loading…
4 tasks done
7
[Model] gemma4 core: quantized inference + gguf loading + fused moe gelu_tanh cpu Related to CPU backends nvidia
#40281 opened Apr 19, 2026 by lesj0610 Loading…
17
[Feat]Make EPLB max expert redundancy configurable
#40280 opened Apr 19, 2026 by nainiu258 Loading…
1
ProTip! Exclude everything labeled bug with -label:bug.