Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,262 Open 15,066 Closed
1,262 Open 15,066 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] Support PCP (prefill context parallel) with MLA needs-rebase v1
#28988 opened Nov 19, 2025 by FENP Loading…
1 of 5 tasks
5
[Feat] Iteration-level profiling for Torch and CUDA profiler nvidia v1
#28987 opened Nov 19, 2025 by benchislett Loading…
2
[Rocm] Set VLLM_ROCM_USE_AITER_FUSION_SHARED_EXPERTS default is disabled rocm Related to AMD ROCm
#28985 opened Nov 19, 2025 by zhyajie Loading…
1
[ROCm][CI] Fix Weight Loading With Multiple GPU Tests on ROCm ci/build rocm Related to AMD ROCm
#28984 opened Nov 19, 2025 by micah-wil Loading…
7
[cuda 13][aarch64][CI] Adding CI steps to build arm64 cuda13 nightly wheels and images aarch64-cuda ci/build nvidia
#28983 opened Nov 19, 2025 by wangshangsam • Draft
5 tasks
@wangshangsam
6
[ROCm][CI] Fixes tests for pytorch nightly and python only builds ci/build rocm Related to AMD ROCm
#28979 opened Nov 19, 2025 by AndreasKaratzas Loading…
6
cleanup at::Tag::needs_fixed_stride_order ready ONLY add when PR is ready to merge/full CI is needed
#28974 opened Nov 19, 2025 by BoyuanFeng Loading…
2
[Feature] add session based streaming support to v1 tpu Related to Google TPUs v1
#28973 opened Nov 19, 2025 by joshuadeng • Draft
2 of 5 tasks
@ywang96
1
[Bugfix] Fix precision loss in LoRA-wrapped RowParallelLinear by fusing bias into GEMM
#28972 opened Nov 19, 2025 by prashanth058 Loading…
2
[LoRA] Support FusedMoE LoRA Triton kernel for mxfp4 model gpt-oss Related to GPT-OSS models
#28971 opened Nov 18, 2025 by xyang16 Loading…
5 tasks
6
Shallow copy SamplingParams and only clone what is needed v1
#28970 opened Nov 18, 2025 by mgoin • Draft
5 tasks
Opt beam search
#28969 opened Nov 18, 2025 by mgoin • Draft
5 tasks
1
[DeepSeek] Fix DeepSeek V3.2 Rope Embedding deepseek Related to DeepSeek models
#28968 opened Nov 18, 2025 by zyongye Loading…
5 tasks
6
[Bug] Fix Batch Invariant MLA test ready ONLY add when PR is ready to merge/full CI is needed v1
#28967 opened Nov 18, 2025 by yewentao256 Loading…
5
Re-enable FlashInfer for Llama4 on Blackwell in e2e fusion tests llama Related to Llama models ready ONLY add when PR is ready to merge/full CI is needed
#28966 opened Nov 18, 2025 by Copilot AI Loading…
3 of 5 tasks
1
@ProExpertProg
2
[Model][QwenVL] Replace torch.repeat_interleave with faster np.repeat qwen Related to Qwen models
#28964 opened Nov 18, 2025 by lgeiger Loading…
3
[Model][QwenVL] Simplify cos/sin rotary embedding indexing qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed
#28962 opened Nov 18, 2025 by lgeiger Loading…
5
[config] Expose get_total_num_hidden_layers() in ModelConfig ready ONLY add when PR is ready to merge/full CI is needed
#28961 opened Nov 18, 2025 by ptovam Loading…
3
[Bugfix] Fix typo in Qwen3 Next model executor qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed
#28960 opened Nov 18, 2025 by Nepherpitou Loading…
2 of 5 tasks
2
[Bugfix] Use lazy string reference for DeepseekV3Config in config registry deepseek Related to DeepSeek models
#28958 opened Nov 18, 2025 by yongming-qin Loading…
8
Update Dockerfile to use gcc-toolset-14 and fix test case failures on power (ppc64le) ci/build v1
#28957 opened Nov 18, 2025 by bhagyashrigai • Draft
5 tasks
[amd] enable kimi k2 fp8 kv on amd mi300 rocm Related to AMD ROCm v1
#28955 opened Nov 18, 2025 by bradleyhd • Draft
5 tasks
Speed up macOS smoke test ci/build ready ONLY add when PR is ready to merge/full CI is needed
#28954 opened Nov 18, 2025 by mgoin Loading…
5 tasks
3
[Core] Free KV cache GPU memory on engine shutdown v1
#28953 opened Nov 18, 2025 by markmc Loading…
5
Relax Transformers modeling backend MoE experts check documentation Improvements or additions to documentation
#28952 opened Nov 18, 2025 by hmellor Loading…
4
ProTip! What’s not been updated in a month: updated:<2025-10-18.