vllm-project/vllm

Pull Requests Commits

[MoE] Add NVIDIA shared expert fusion

WoosukKwon committed 28 days ago

6b6d7d02

[Bugfix] Import each packed IPC export once on the consumer side (#51259)

acmore committed 28 days ago

Verified f37dff22

[Frontend] Add content_parts to /inference/v1/generate for raw multim… (#51478)

aoshen02 committed 28 days ago

Verified 33584901

[CI] Solidify speculative decoding E2E coverage (#50713)

AndreasKaratzas committed 28 days ago

Verified 1a172733

Fix chat completion 500 on non-object JSON bodies (#51654)

tarukumar committed 28 days ago

Verified 48bada6e

[Rust Frontend] Support dynamic tools from developer messages (#51144)

BugenZhao committed 28 days ago

Verified f1e921b6

[MM][CG][BugFix] Fix Ernie-4.5-VL encoder CG postprocess for multi-path outputs (#51461)

qyYue1389 committed 29 days ago

Verified b2506d62

[Rust Frontend] Upgrade MiniJinja to 2.22 & remove method lookup workaround (#51235)

BugenZhao committed 29 days ago

Verified d8c70f22

[CI] Parallelize release image publishing (#51735)

khluu committed 29 days ago

Verified 8a9f9f76

[Perf] Skip detokenization in offline beam search (#50333)

samuelkim7 committed 29 days ago

Verified 8977ea88

[Bugfix] Fix DeepSeek V4/3.2 tokenizer vocab size overcount crashing guided decoding (#51727)

sfeng33 committed 29 days ago

Verified 8bcc916a

[Model Runner V2][MTP] Share topk index buffer between draft steps (#47352)

TheEpicDolphin committed 29 days ago

Verified 3e174bb7

[Bugfix][MiMo] Apply vision attention sinks in the window attention path (#49815)

almogtavor committed 29 days ago

Verified c3cac8c6

[Bugfix] Preserve non-logitproc entry points in tests (#51097)

d4l3k committed 29 days ago

Verified 98a4144a

[Rust Frontend][gRPC] Add explicit data-parallel rank routing (#51178)

connorcarpenter15 committed 29 days ago

Verified fa722b9f

[CI] Add /ci cancel command (#51732)

khluu committed 29 days ago

Verified 3c6b2e9c

[Build][gRPC] Publish protobuf schemas to Buf (#51276)

connorcarpenter15 committed 29 days ago

Verified 635dd6aa

[Kimi-K3] DCP support (#50484)

GirasoleY committed 29 days ago

Verified 63ac04a6

[Bugfix] Align deepseek v4 parser thinking default with tokenizer (#51296)

sfeng33 committed 29 days ago

Verified 3f142bd8

Fix DSpark warmup without sparse index buffer (#50693)

xijiaat committed 29 days ago

Verified d40c3e3c

[ROCm][DistInf] Enable vLLM DI CI with buildkite/slurm (#47030)

lcskrishna committed 29 days ago

Verified 1482d2e0

[Bugfix][Structured Output] Mask request stop tokens in xgrammar until grammar terminates (#49227)

sfeng33 committed 29 days ago

Verified 05f0a800

[Perf] Launch the top-k/top-p Triton sampler kernel with 8 warps (#51507)

BabyDrangoner committed 29 days ago

Verified 405bc867

[Perf] Narrow DeepSeek V4 eager CUDA graph region (#51430)

WoosukKwon committed 29 days ago

Verified 79c865b8

[Docker] Cache test dependencies before vLLM install (#51184)

mgoin committed 29 days ago

Verified 21c667aa

[BugFix][SpecDecode] Fix dspark parallel_drafting_token_id init bug (#51602)

wangxiyuan committed 29 days ago

Verified 355a338b

[Build] Skip precompiled wheel fetch during metadata hooks (#51424)

mgoin committed 29 days ago

Verified 3a749ce8

[Bugfix][Kimi-K3] Give the AMD packed KDA decode kernel the state-index stride (#51682)

xudonlyu committed 29 days ago

Verified 0e2d7802

Fix DoS via sample-rate forgery bypassing audio decode duration guard (#49948)

jperezdealgaba committed 29 days ago

Verified 640a0908

[ROCm][MLA] [K3] Fix fp8 KV cache decode on the AITER MLA backend (#51011)

fanxingran committed 29 days ago

Verified 37fbf520

Older