Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,164 Open 13,701 Closed
1,164 Open 13,701 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

Notice for deprecation of AutoAWQ documentation Improvements or additions to documentation
#26820 opened Oct 14, 2025 by HDCharles Loading…
4 tasks done
5
AI Fix for: [Feature]: Add process_weights_after_loading to AttentionImpl frontend
#26819 opened Oct 14, 2025 by shanaya-Gupta Loading…
1
7
[Kernel][MoE] Add MoE tunings for GLM 4.6-FP8 and GLM 4.5 Air on NVidia B200
#26818 opened Oct 14, 2025 by zklapow Loading…
3 of 5 tasks
3
[CI Failure] Fix tests with missing TinyLlama-1.1B-Chat-v1.0-FP8-e2e llama Related to Llama models ready ONLY add when PR is ready to merge/full CI is needed
#26816 opened Oct 14, 2025 by mgoin Loading…
5 tasks
4
[Bugfix] Fix qwen3-omni audio truncation issue qwen Related to Qwen models
#26815 opened Oct 14, 2025 by Isotr0py Loading…
1 of 5 tasks
1
2
[P/D] KV Load Failure Recovery/Abort Configuration frontend kv-connector v1
#26813 opened Oct 14, 2025 by wseaton Loading…
4
[Nixl] Add metrics to Prometheus-Grafana dashboard kv-connector v1
#26811 opened Oct 14, 2025 by NickLucche Loading…
[Core] Use envs.__getattr__ for all Unify to environment variable access multi-modality Related to multi-modality (#4194) v1
#26810 opened Oct 14, 2025 by Jialin Loading…
3 of 5 tasks
1
[Docs] update README.md to display logo correctly and fix links documentation Improvements or additions to documentation
#26809 opened Oct 14, 2025 by ddalgrande Loading…
3 of 5 tasks
8
[Feature] GatedDeltaNet Automatic Prefix Caching qwen Related to Qwen models v1
#26807 opened Oct 14, 2025 by simondanielsson • Draft
1 of 11 tasks
3
Fix seed reproducibility issue by adding output.copy_(out)
#26805 opened Oct 14, 2025 by XuanofXXX Loading…
3 of 5 tasks
4
[wip] Allow oot custom compiler extension via CompilerInterface needs-rebase
#26804 opened Oct 14, 2025 by wxsIcey • Draft
5 tasks
1
[wip] tools for slo management frontend v1
#26802 opened Oct 14, 2025 by hj-mistral • Draft
5 tasks
9
[Metrics] Refactor LoRA state tracking ready ONLY add when PR is ready to merge/full CI is needed v1
#26801 opened Oct 14, 2025 by markmc Loading…
4
[Model] add kosmos2_5 for vllm new-model Requests to new models
#26800 opened Oct 14, 2025 by yugeeklab Loading…
3 of 5 tasks
8
[UX] Fallback to native implementation when flashinfer sampler failed to compile v1
#26799 opened Oct 14, 2025 by Isotr0py Loading…
1 of 5 tasks
4
[V1][performance] add multi step v1
#26796 opened Oct 14, 2025 by chengda-wu Loading…
5 tasks
6
[Doc] ruff format remaining Python examples documentation Improvements or additions to documentation
#26795 opened Oct 14, 2025 by DarkLight1337 Loading…
5 tasks
3
make fp4 scaled_mm works for 5090 gpu ci/build
#26793 opened Oct 14, 2025 by XiaobingSuper Loading…
3 of 5 tasks
3
llama4_vision_rope: add HIP override to accept (q, k) and avoid (positions, q, k) mismatch llama Related to Llama models
#26790 opened Oct 14, 2025 by hl475 Loading…
5 tasks
[bugfix] remove unused parameters to reduce unnecessary vram usage ready ONLY add when PR is ready to merge/full CI is needed
#26789 opened Oct 14, 2025 by ReinForce-II Loading…
3 of 5 tasks
4
[Feature] default --extra-body param to disable thinking in vllm bench serve frontend performance Performance-related issues ready ONLY add when PR is ready to merge/full CI is needed
#26784 opened Oct 14, 2025 by lengrongfu Loading…
5 tasks
1
9
[Fix] Avoid UserWarning when creating tensors from base64 embeddings documentation Improvements or additions to documentation
#26782 opened Oct 14, 2025 by mmangkad Loading…
5 tasks
1
9
[Bugfix] DeepSeek V3.2 MTP metadata & CUDA graph issues deepseek Related to DeepSeek models speculative-decoding v1
#26779 opened Oct 14, 2025 by xiaohajiayou Loading…
3
[CI/Build][Bugfix] fix qutlass cmake error when set QUTLASS_SRC_DIR bug Something isn't working ci/build ready ONLY add when PR is ready to merge/full CI is needed
#26773 opened Oct 14, 2025 by izhuhaoran Loading…
5 tasks
5
ProTip! Exclude everything labeled bug with -label:bug.