Skip to content

Pull requests: vllm-project/vllm

New pull request New
3,762 Open 27,583 Closed
3,762 Open 27,583 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] Reduce NVFP4 MoE load memory under sleep mode bug Something isn't working nvidia
#48701 opened Jul 15, 2026 by hubunt • Draft
1
2
[Bugfix]Fix transformer backend failed: AttributeError: 'Parameter' object has no attribute 'weight_loader' bug Something isn't working
#48699 opened Jul 15, 2026 by Yejing-Lai Contributor Loading…
[Bugfix] Scale FlashInfer b12x capacity for data parallelism bug Something isn't working nvidia
#48698 opened Jul 15, 2026 by dumko2001 Loading…
3 of 4 tasks
1
4
[Bugfix] Reject DeepSeek V4 FP4 MoE tensor parallelism bug Something isn't working deepseek Related to DeepSeek models
#48697 opened Jul 15, 2026 by dumko2001 Loading…
3 of 4 tasks
1
4
[Bugfix] Pad misaligned MXFP4 MoE shards for Marlin bug Something isn't working
#48696 opened Jul 15, 2026 by dumko2001 Loading…
3 of 4 tasks
1
4
[Bugfix] Gate FlashInfer FP4 MoE on kernel availability bug Something isn't working nvidia
#48695 opened Jul 15, 2026 by dumko2001 Loading…
3 of 4 tasks
1
4
[Bugfix] Avoid FlashAttention 3 with batch invariance on Hopper bug Something isn't working v1
#48694 opened Jul 15, 2026 by dumko2001 Loading…
3 of 4 tasks
1
4
[Bugfix] Make Gemma 4 suppress-token masking CUDA-graph safe bug Something isn't working nvidia
#48693 opened Jul 15, 2026 by dumko2001 Loading…
3 of 4 tasks
1
4
[MRV2][Spec Decode] Adaptive Speculative Decoding - Initial Support documentation Improvements or additions to documentation nvidia qwen Related to Qwen models speculative-decoding v1
#48692 opened Jul 15, 2026 by benchislett Member Loading…
4 tasks done
1
[Spec Decode] Add support for num draft layers > 1 for Eagle speculative-decoding v1
#48691 opened Jul 15, 2026 by ekagra-ranjan Contributor • Draft
1 of 2 tasks
[Bugfix] MiniMax M3 reasoning parser: treat non-leading <mm:think> as content bug Something isn't working tool-calling
#48690 opened Jul 15, 2026 by harjothkhara Loading…
1
1
[ROCm][Bugfix] Enable the fp32 head_dtype torch.mm fast path on ROCm bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm v1 verified Run pre-commit for new contributors without triggering other tests
#48688 opened Jul 15, 2026 by wjabbour Contributor Loading…
1
[chore] code refactor v1
#48687 opened Jul 15, 2026 by andyxning Contributor Loading…
4 tasks
[Model] Add minimal native RWKV7 serving support documentation Improvements or additions to documentation new-model Requests to new models
#48686 opened Jul 15, 2026 by shiroko98 • Draft
2
[Benchmarks] Fix type hints and argparse help texts performance Performance-related issues structured-output
#48685 opened Jul 15, 2026 by whygyc Loading…
3 of 4 tasks
1
[Misc] Remove override_attention_dtype
#48684 opened Jul 15, 2026 by wangxiyuan Contributor Loading…
4 tasks
[ROCm] Bump AITER to v0.1.16.post4 ci/build rocm Related to AMD ROCm
#48683 opened Jul 15, 2026 by Fangzhou-Ai Contributor Loading…
4 tasks done
1
[Bugfix] Fix HYV3 shared_mlp prefix for compressed-tensors ignore matching bug Something isn't working
#48682 opened Jul 15, 2026 by DCoEngine Loading…
3
[XPU][CI] Split V1 Sample + Logits CI job into two parallel steps ci/build intel-gpu Related to Intel GPU
#48681 opened Jul 15, 2026 by zhenwei-intel Contributor Loading…
4 tasks
[KV Offload] Add self-describing events for tier promotions documentation Improvements or additions to documentation kv-connector v1
#48679 opened Jul 15, 2026 by Change72 Contributor • Draft
1
[Bugfix] Gemma4 parser: don't truncate string args at interior <|"|> tokens bug Something isn't working tool-calling
#48678 opened Jul 15, 2026 by troymroberts Loading…
1
1
[XPU] upgrade to torch 2.13 ci/build intel-gpu Related to Intel GPU
#48677 opened Jul 15, 2026 by yma11 Contributor • Draft
4 tasks
3
[ROCm][CI] fix test_common.py ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
#48676 opened Jul 15, 2026 by charlifu Contributor Loading…
2
fix(v1): freeze offline multiprocess front-end heap (#48229) frontend v1
#48675 opened Jul 15, 2026 by aryanyadav0402 Loading…
1
[Bugfix] Fix logprobs token-string collision from SentencePiece space… bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed
#48674 opened Jul 15, 2026 by aoshen02 Collaborator Loading…
4 tasks
1
@chaunceyjiang
2
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.