Skip to content

Pull requests: vllm-project/vllm

New pull request New
2,775 Open 22,825 Closed
2,775 Open 22,825 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

[Spec Decode] Don't compile update_num_computed_tokens_for_batch_change speculative-decoding v1
#41484 opened May 1, 2026 by MatthewBonanni Collaborator • Draft
4 tasks
1
2
[bug] Fix lora reloading from disk every step when using AsyncLLM bug Something isn't working frontend
#41482 opened May 1, 2026 by hao-aaron Contributor Loading…
4 tasks
[Spec Decode][V1][V2] Warm spec-decode helper kernels at startup speculative-decoding v1
#41481 opened May 1, 2026 by li-xinwei Loading…
1
5
[Bugfix] Fix B200 batch determinism in fused_moe logic bug Something isn't working
#41480 opened May 1, 2026 by Lucaskabela Contributor • Draft
3 of 4 tasks
1
6
[Bug] Fix /pooling 404 after resolve_lora in _maybe_get_adapters bug Something isn't working frontend
#41479 opened May 1, 2026 by MattThomas-fastino Loading…
4
Limit concurrency on test_transcription_api_correctness.py
#41478 opened May 1, 2026 by ekagra-ranjan Contributor Loading…
4
[Perf] Batch Weight Prefetching via cuMemcpyBatchAsync to Reduce Latency
#41474 opened May 1, 2026 by xiaobao520123 Loading…
3 of 4 tasks
7
[WIP] Clean up scheduling loop v1
#41473 opened May 1, 2026 by andylolu2 Contributor • Draft
3
[Refactor] Remove dead code in tests and parallel_state kv-connector ready ONLY add when PR is ready to merge/full CI is needed v1
#41471 opened May 1, 2026 by yewentao256 Member Loading…
[Bugfix] Validate 'text' field in Responses API multi-modal content bug Something isn't working frontend
#41470 opened May 1, 2026 by ankrovv Loading…
4 tasks done
4
[Bugfix] Detect MTP truncation at reasoning-to-tool-call boundary bug Something isn't working frontend
#41467 opened May 1, 2026 by ToastyTheBot • Draft
4 tasks
5
[Bugfix] Fix Qwen3Coder prev_tool_call_arr double-emission on parse failure bug Something isn't working qwen Related to Qwen models tool-calling
#41466 opened May 1, 2026 by ToastyTheBot • Draft
3 tasks
5
[BugFix][MyPy]: Module has no attribute "sched_getaffinity" [attr-defined] bug Something isn't working cpu Related to CPU backends
#41465 opened May 1, 2026 by hickeyma Contributor Loading…
Fix DeepSeek-OCR for Transformers v4 deepseek Related to DeepSeek models
#41460 opened May 1, 2026 by hmellor Member Loading…
6
fix(frontend): Add multimodal placeholders to Gemma4 tool message template documentation Improvements or additions to documentation
#41459 opened May 1, 2026 by harshaljanjani Loading…
4 of 5 tasks
1
2
[Perf] Fuse Qwen3.5 GDN in_proj_ba into 6-way in_proj MergedColumnParallelLinear qwen Related to Qwen models
#41457 opened May 1, 2026 by jhsmith409 Contributor Loading…
5 tasks done
6
[ROCM][RDNA3] WMMA paged prefill and split-K decode kernels for ROCM_ATTN ci/build rocm Related to AMD ROCm v1
#41455 opened May 1, 2026 by JartX Contributor Loading…
13
[Bugfix] Reject unsupported FlexAttention head sizes bug Something isn't working v1
#41454 opened May 1, 2026 by bugkeep Loading…
5
[CI] Route part of B200 jobs to b200-k8s ci/build ready ONLY add when PR is ready to merge/full CI is needed
#41453 opened May 1, 2026 by khluu Collaborator • Draft
3
[ROCm][Deepseekv4] DeepseekV4 Mi300 support ci/build deepseek Related to DeepSeek models documentation Improvements or additions to documentation rocm Related to AMD ROCm v1
#41451 opened May 1, 2026 by ganyi1996ppo Contributor • Draft
4 tasks
5
fix: handle missing parent modules in _has_module
#41449 opened May 1, 2026 by ShuhaoZhangTony Loading…
3
[Kernel][AMD] Optimize GatedDeltaNet FLA prefill kernels on MI300X rocm Related to AMD ROCm
#41446 opened May 1, 2026 by zobinHuang Loading…
1
[DSV4] AR+mhc_post fusion ci/build nvidia performance Performance-related issues
#41441 opened May 1, 2026 by jeejeelee Collaborator • Draft
4 tasks
6
Revert "[MoE] Make MoERunnerInterface a PluggableLayer for OOT support" (#35178) documentation Improvements or additions to documentation
#41440 opened May 1, 2026 by vllm-agent • Draft
7
Revert "Fix Cohere ASR after HF upgrade" (#40582) multi-modality Related to multi-modality (#4194)
#41439 opened May 1, 2026 by vllm-agent • Draft
1
ProTip! Updated in the last three days: updated:>2026-04-28.