Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,350 Open 17,059 Closed
1,350 Open 17,059 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

[Attention
#31851 opened Jan 7, 2026 by LucasWilkinson Loading…
5 tasks
[Attention][3/n] Remove usage of deprecated seq_lens_cpu and num_computed_tokens_cpu CommonAttentionMetadata properties ready ONLY add when PR is ready to merge/full CI is needed v1
#31850 opened Jan 7, 2026 by LucasWilkinson Loading…
2
[CI] Fix weight mapping test for transformers v5 tied weights multi-modality Related to multi-modality (#4194)
#31849 opened Jan 7, 2026 by AndreasKaratzas Loading…
1
[Model] Add Grok-2 documentation Improvements or additions to documentation
#31847 opened Jan 7, 2026 by dangoldbj Loading…
5 tasks
12
Optimize graph capture size
#31846 opened Jan 7, 2026 by jiahanc • Draft
5 tasks
2
[4/n] Migrate pos_encoding sampler and fused_qknorm_rope to libtorch stable ABI ci/build cpu Related to CPU backends nvidia
#31842 opened Jan 6, 2026 by mikaylagawarecki • Draft
5 tasks
8
[Bugfix] Fix race condition in async-scheduling for vlm model v1
#31841 opened Jan 6, 2026 by tianshu-Michael-yu Loading…
3 of 5 tasks
1
[responsesAPI] get reasoning token metrics for simpleContext frontend gpt-oss Related to GPT-OSS models
#31839 opened Jan 6, 2026 by qandrew • Draft
5 tasks
3
[1/2][lmcache connector] clean up lmcache multi-process adapter kv-connector ready ONLY add when PR is ready to merge/full CI is needed
#31838 opened Jan 6, 2026 by ApostaC Loading…
5 tasks
2
[Perf] Fuse stride preparation for NVFP4 cutlass_moe nvidia performance Performance-related issues ready ONLY add when PR is ready to merge/full CI is needed
#31837 opened Jan 6, 2026 by mgoin Loading…
5 tasks
[responsesAPI] fix incomplete_messages for simple/parsable context frontend
#31836 opened Jan 6, 2026 by qandrew Loading…
3
[CI/Build] Enable test_kv_cache_events_dp for AMD rocm Related to AMD ROCm v1
#31834 opened Jan 6, 2026 by rjrock Loading…
3 tasks done
3
[ROCm][CI] v1 cpu offloading attention backend fix rocm Related to AMD ROCm v1
#31833 opened Jan 6, 2026 by AndreasKaratzas Loading…
5
[Perf][Kernel] Fused SiLU+Mul+Quant kernel for NVFP4 cutlass_moe nvidia performance Performance-related issues
#31832 opened Jan 6, 2026 by mgoin Loading…
5 tasks
3
[Perf] Optimize cutlass moe problem size calculation, 5.3% E2E Throughput improvement, 2.2% TTFT improvement nvidia ready ONLY add when PR is ready to merge/full CI is needed
#31830 opened Jan 6, 2026 by yewentao256 Loading…
3
[Perf] Add opt-in SM100 Oink RMSNorm custom-op path
#31828 opened Jan 6, 2026 by Laurawly Loading…
4
[MoE Refactor][17/N] Apply Refactor to Bf16
#31827 opened Jan 6, 2026 by zyongye Loading…
5 tasks
10
[Perf] Slight improvement of ITL with multiple GPUs
#31826 opened Jan 6, 2026 by access2rohit Loading…
2 of 5 tasks
4
[Model] Enable LoRA support for tower and connector in DotsOCR documentation Improvements or additions to documentation
#31825 opened Jan 6, 2026 by ShaanveerS Loading…
3
[CI] Add CUDA 13 nightly containers ci/build nvidia
#31822 opened Jan 6, 2026 by csahithi Loading…
5 tasks
3
[ROCm][AITER] bugfix accuracy regression in ROCM_AITER_TRITON_MLA backend rocm Related to AMD ROCm v1
#31816 opened Jan 6, 2026 by vllmellm Loading…
5 tasks
[Bugfix] Fix TorchAO quantization bugs and add --torchao-config CLI support
#31815 opened Jan 6, 2026 by jwpark33 Loading…
5 tasks
1
3
[Bugfix] Inject JSON schema descriptions into prompt for structured outputs frontend
#31814 opened Jan 6, 2026 by ricky-chaoju Loading…
1
3
Enable LoRA support for tower and connector in Mistral and Voxtral deepseek Related to DeepSeek models documentation Improvements or additions to documentation frontend qwen Related to Qwen models
#31812 opened Jan 6, 2026 by Anexdeus Loading…
5
[KVConnector] Handle KV events from multiple connectors kv-connector v1
#31811 opened Jan 6, 2026 by hickeyma • Draft
3 of 5 tasks
6
ProTip! Exclude everything labeled bug with -label:bug.