Skip to content

Pull requests: vllm-project/vllm

New pull request New
4,094 Open 28,510 Closed
4,094 Open 28,510 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[KV Offload] Enable single-copy MLA layout for CPUOffloadingSpec v1
#50301 opened Jul 29, 2026 by Change72 Contributor • Draft
[Core] Add TP-invariant tree kernels across TP sizes
#50299 opened Jul 29, 2026 by cm2435 • Draft
1
[Security] Fix chat template resource-exhaustion DoS (GHSA-4hhp-h66f-… frontend rust
#50300 opened Jul 29, 2026 by jperezdealgaba Contributor • Review required
6
[DSv4 Perf] Remove redundant full kernel for dsv4, 1.88x kernel performance improvement ready ONLY add when PR is ready to merge/full CI is needed
#50298 opened Jul 29, 2026 by yewentao256 Member • Review required
[BugFix] Fix P/D preemption race condition bug Something isn't working kv-connector ready ONLY add when PR is ready to merge/full CI is needed v1
#50297 opened Jul 29, 2026 by njhill Member • Review required
2
[Test] Add per-parser tool-call format conformance tests tool-calling
#50296 opened Jul 29, 2026 by guanxingithub • Review required
2
[Kernel][Model] Optimize FA4 mm_prefix range lookup v1
#50294 opened Jul 29, 2026 by vhagor • Review required
1
[Model Runner V2] Enable encoder token classification ready ONLY add when PR is ready to merge/full CI is needed v1
#50293 opened Jul 29, 2026 by taneem-ibrahim Contributor • Approved
2
[Bugfix] Preserve truncated Kimi K3 reasoning bug Something isn't working k3 kimi tool-calling
#50291 opened Jul 29, 2026 by rohitsalla • Draft
1
1
[Model] Step3: guard tie_word_embeddings access for pipeline parallelism
#50290 opened Jul 29, 2026 by chanh Contributor • Draft
[Rust Frontend] Add standalone Rust renderer rust
#50289 opened Jul 29, 2026 by sagearc Contributor • Draft
[SM120] Add NVFP4 KV cache support for consumer Blackwell (RTX 5090) ci/build nvidia v1
#50288 opened Jul 29, 2026 by ch2lab • Review required
4 tasks
1
[Bugfix][Model Runner V2] Preserve Mamba block table capacity under DCP bug Something isn't working v1
#50287 opened Jul 29, 2026 by jongukc Contributor • Review required
3 of 4 tasks
[Refactor] Remove multiple dead codes frontend quantization ready ONLY add when PR is ready to merge/full CI is needed v1
#50285 opened Jul 29, 2026 by yewentao256 Member • Review required
[CI] Stabilize speculator memory teardown ready ONLY add when PR is ready to merge/full CI is needed v1
#50284 opened Jul 29, 2026 by AndreasKaratzas Member • Review required
[Core] Add --enable-nan-fault-tolerance for NaN detection and request abort v1
#50283 opened Jul 29, 2026 by elvircrn Contributor • Draft
6 tasks
[Bugfix][Spec Decode] Make EAGLE weight sharing TP-consistent bug Something isn't working speculative-decoding v1
#50280 opened Jul 29, 2026 by sdflkjssl • Review required
1
1
[Kernel] Fuse strided GDN RMSNorm gating on SM103 performance Performance-related issues
#50279 opened Jul 29, 2026 by Alizen-1009 • Review required
2
[Bugfix][LoRA] Make Punica shrink deterministic with split_k=1 bug Something isn't working
#50278 opened Jul 29, 2026 by fooSynaptic • Review required
3 of 4 tasks
1
2
[Bugfix] Make Rust HF dataset endpoint configurable bug Something isn't working rust
#50277 opened Jul 29, 2026 by rohitsalla • Draft
1
1
[Bugfix] Fix packed KV block zeroing stride bug Something isn't working v1 verified Run pre-commit for new contributors without triggering other tests
#50276 opened Jul 29, 2026 by wangxian001 • Review required
10
[Bugfix][EC Connector] Don't stop an encoder-instance request before its images are encoded bug Something isn't working v1
#50275 opened Jul 29, 2026 by gty111 Contributor • Review required
4 tasks
2
Fix Qwen Omni timing for presampled videos multi-modality Related to multi-modality (#4194) qwen Related to Qwen models
#50274 opened Jul 29, 2026 by ccyyy1023 • Review required
1
1
[Quantization] Honor --linear-backend for ModelOpt W4A16 quantization ready ONLY add when PR is ready to merge/full CI is needed
#50273 opened Jul 29, 2026 by netanel-haber Contributor • Approved
1
[Bugfix] Fix speculative decoding for short_conv (LFM2) models bug Something isn't working v1
#50272 opened Jul 29, 2026 by zwischenraum • Review required
1
ProTip! What’s not been updated in a month: updated:<2026-06-29.