Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,656 Open 18,873 Closed
1,656 Open 18,873 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

[WIP][Bugfix] Fix MLA attention crash with AWQ/GPTQ quantized models bug Something isn't working
#34695 opened Feb 17, 2026 by haosdent Loading…
1
[Bugfix] Fix mypy errors for StructuredOutputsParams by using stdlib dataclass bug Something isn't working
#34693 opened Feb 17, 2026 by hyeongyun0916 Loading…
3 of 5 tasks
[ROCm] Enable DeepEP ROCm as all2allbackend for AMD GPUs. rocm Related to AMD ROCm
#34692 opened Feb 17, 2026 by lcskrishna • Draft
5 tasks
4
[WIP][Bugfix] Fix xgrammar nanobind leaked objects at shutdown bug Something isn't working structured-output v1
#34690 opened Feb 17, 2026 by haosdent Loading…
1
2
[ROCm] Enable bitsandbytes quantization support on ROCm ci/build documentation Improvements or additions to documentation ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
#34688 opened Feb 17, 2026 by Abdennacer-Badaoui Loading…
1 task done
5
[Update] Use FlashInfer fast_decode_plan directly instead of replication nvidia v1
#34687 opened Feb 17, 2026 by askliar Loading…
1
[WIP][Bugfix] Fix illegal memory access in causal_conv1d Triton kernels with CUDA graphs bug Something isn't working nvidia
#34685 opened Feb 17, 2026 by haosdent Loading…
1
2
[kv_offload+HMA][2/N]: Support sliding window lookup kv-connector v1
#34682 opened Feb 17, 2026 by orozery Loading…
5
[kv_offload+HMA][1/N]: Worker-side support for multiple HMA groups v1
#34680 opened Feb 17, 2026 by orozery Loading…
[Docs]Fix documentation formatting in architecture overview documentation Improvements or additions to documentation
#34679 opened Feb 17, 2026 by lichuang Loading…
5 tasks
2
[GGUF][Model] Add Qwen3-Coder-Next GGUF support multi-modality Related to multi-modality (#4194) qwen Related to Qwen models
#34678 opened Feb 17, 2026 by rudybear Loading…
6 tasks done
4
[Bugfix][CPU] Fix basic unit tests failing in CPU platforms bug Something isn't working nvidia
#34677 opened Feb 17, 2026 by jasonyanwenl Loading…
3 of 5 tasks
3
Add VLLM_SKIP_MODEL_VALIDATION environment variable frontend
#34676 opened Feb 17, 2026 by dsingal0 Loading…
5 tasks
3
Adding Nemotron fp8 Triton MoE Config
#34674 opened Feb 17, 2026 by yugong333 Loading…
5 tasks
2
[Bugfix][MOE] Fix incorrect routing selection for models without expert groups (e.g., MiniMax-M2.1) bug Something isn't working nvidia
#34673 opened Feb 17, 2026 by wwl2755 Loading…
1
[ci] Add Ray compatibility check informational CI job ci/build
#34672 opened Feb 17, 2026 by jeffreywang-anyscale • Draft
2 of 7 tasks
3
Update max_num_tokens value when specdec is enabled v1
#34671 opened Feb 17, 2026 by shaharmor98 Loading…
5 tasks
2
[Core] Default to "mp" rather than "uni" distributed executor backend ready ONLY add when PR is ready to merge/full CI is needed v1
#34670 opened Feb 17, 2026 by njhill Loading…
4
[Reasoning] [Draft][WIP] Support for speculative decoding with thinking budget frontend v1
#34668 opened Feb 17, 2026 by rishitdholakia13 • Draft
5
[Bugfix] Fix benchmark_fused_collective crash on CustomOp init bug Something isn't working performance Performance-related issues
#34665 opened Feb 17, 2026 by mayank-ketkar-sf Loading…
3 tasks done
5
Add MXFP8 to Marlin dense kernel
#34664 opened Feb 17, 2026 by mgoin Loading…
5 tasks
2
Separate TRTLLM and Flashinfer backends documentation Improvements or additions to documentation nvidia v1
#34663 opened Feb 17, 2026 by pavanimajety • Draft
5 tasks
3
[Kernel][Perf] Fuse gather_block_tables + compute_slot_mappings into single kernel performance Performance-related issues v1
#34660 opened Feb 17, 2026 by mayank-ketkar-sf Loading…
3 of 5 tasks
9
Refactor oracle to separate support and selection needs-rebase
#34658 opened Feb 16, 2026 by mgoin • Draft
5 tasks
3
[Quant][WIP] Add compressed-tensors MXFP4 support through flashinfer nvidia
#34657 opened Feb 16, 2026 by dsikka • Draft
5 tasks
4
ProTip! Mix and match filters to narrow down what you’re looking for.