vllm-project/vllm

Pull Requests Commits

[Build] Pin DeepGEMM CUDA FP8 header fix

khluu committed 6 days ago

00da1d78

[Bugfix][Build] Restore DeepGEMM SM120 support with SITU

ZacharyZcR committed 7 days ago

130637ee

[Bugfix][KV Connector] Propagate EAGLE state across merged Mooncake store groups (#49069)

ivanium committed 7 days ago

Verified 5c4fe4b1

[Bugfix] serving_llama70B_tp4 benchmark was silently running at tensor_parallel_size=1 (#50766)

wjabbour committed 7 days ago

Verified e42c230e

[Model] Support jina-embeddings-v5-text-nano (EuroBERT encoder backbone) (#50688)

omkar-droid committed 7 days ago

Verified 9a4fd57c

[Test][V1] Add sleep/wake correctness regression test for hybrid GDN/… (#44972)

chun-wan committed 7 days ago

Verified 0033211c

[Attention][MiniMax-M3] Add MSA speculative decode verification (#50032)

jasonlizhengjian committed 7 days ago

Verified 0055b8bf

[ROCm][Bugfix][Kimi-K3] Preserve MoE correction bias in FP32 (#50761)

Fangzhou-Ai committed 7 days ago

Verified c6668106

[Model Runner v2] Enable BGE M3 pooling embed token_classify (#50661)

taneem-ibrahim committed 7 days ago

Verified 55c98e37

[Elastic EP] Fix non-contiguous weight transfers (#50641)

itayalroy committed 7 days ago

Verified 96add737

[Bugfix][Models] Accept Qwen3_5MoeTextConfig in Qwen3_5MoeProcessingInfo for transformers 5.x compatibility (#50704)

loulanyue committed 7 days ago

Verified 06018507

[1/N] Unify multiple-path encoder cuda graph support (#49934)

Isotr0py committed 8 days ago

Verified e2fa2859

[Bugfix][Doc] Fix references to FusedMoE in doc (#50701)

bnellnm committed 8 days ago

Verified c67fe497

[DSV4] Implement Sequence Parallelism (#46789)

WoosukKwon committed 8 days ago

Verified 38a466e7

[GPT-OSS] Strict tool call and constrained decoding for Harmony (#45560)

yzong-rh committed 8 days ago

Verified dc818c19

Add @hongxiayang as code owner for AMD-specific model files and ROCm docs (#50608)

hongxiayang committed 8 days ago

Verified 127a7ceb

[model registry] some simple typos (#50673)

andyxning committed 8 days ago

Verified 03c782eb

[Core] Offload raw-prompt preprocessing to renderer thread pool in AsyncLLM (#49608)

almogtavor committed 8 days ago

Verified 39f55ffd

[Benchmark] Add probe requests to vllm bench serve (#49611)

guan404ming committed 8 days ago

Verified 63e78ce3

[Bugfix][Kernel] Fix dangling temporary in AWQ gemm torch::stable::sum dim arg (#46805)

wentian-byte committed 8 days ago

Verified ab064861

[Model Runner V2] Enable encoder token embedding (#50574)

taneem-ibrahim committed 8 days ago

Verified 652ba592

(feat): optionally disable lookup on PD decode (#50498)

majunze2001 committed 8 days ago

Verified 3986b967

[Frontend] Add cache_salt support to Anthropic Messages API (#49498)

aeon-x committed 8 days ago

Verified 81a42d37

[ROCm][MLA] Mask the AITER MLA small-head verify flatten causally (#50476)

yudigege86 committed 8 days ago

Verified 77469c90

Add Understanding the Latency Metrics docs (#50600)

mgoin committed 8 days ago

Verified 4ee9702b

[Frontend] Cohere chat v2 api support (#47189)

andrewbcohere committed 8 days ago

Verified 9c110fa5

[Bugfix][Parser] Forward model_config to nested reasoning parsers (#50642)

chaunceyjiang committed 9 days ago

Verified 6c91de36

Add @shen-shanshan to CODEOWNERS (#50655)

shen-shanshan committed 9 days ago

Verified 124154a8

[Bugfix][CI] Prevent common ops imports from initializing CUDA (#50639)

AndreasKaratzas committed 9 days ago

Verified f7097a92

[Rust][Benchmark] Prevent invalid token IDs in random benchmarks (#50058)

reidliu41 committed 9 days ago

Verified 62195e97

Older