Commits vllm-project/vllm

fix ci pre-commit

yewentao256 committed 17 days ago

7d98b72b

[Core] Update PyTorch to 2.13.0, torchvision to 0.28.0, triton to 3.7.1 (#48155)

atalman committed 17 days ago

Verified 75ccdf31

[CI] Bump PyTorch Compilation Unit Tests timeout to 150 min (#49606)

atalman committed 17 days ago

Verified c6fe94b4

[CI][Bugfix] Fix test isolation in block_int8/ptpc_fp8 MoE kernel tests (#49609)

njhill committed 17 days ago

Verified 46f01a50

[CI] Isolate cudagraph tests in child processes (#49510)

AndreasKaratzas committed 17 days ago

Verified f00efc52

[DSv4 Perf] Skip topk and router when not needed, 3.4% E2E TTFT improvement for Decode case (#49486)

yewentao256 committed 17 days ago

Verified b0cb1da1

[CI] Use explicit devices in quantization tests (#49512)

AndreasKaratzas committed 17 days ago

Verified 0e36e3bb

Revert "[MRV2] Always build attn metadata at capture time" (#49364) (#49451)

vllm-agent committed 17 days ago

Verified 494845e7

[Bugfix][Structured Output][Spec Decode] Advance grammar across reasoning boundary (#44993)

yuyue0225sc committed 17 days ago

Verified 0416dab2

Fix GPTQ quantized Qwen3.5 MTP weight loading with spec decode (#48816)

vllmellm committed 17 days ago

Verified c8db00b1

[Perf] Defer MM embeds loading off the event loop (#49477)

guan404ming committed 17 days ago

Verified 80c76839

[Bugfix][CI/Build] Fix Plamo2 HF runner crash on transformers v5 (_tied_weights_keys list→dict) (#44239)

nikhilkulkarni1755 committed 17 days ago

Verified 638d6e97

[Bugfix][Spec Decode] Select earliest-completing stop string in check_stop_strings (#49391)

davidjpyu committed 17 days ago

Verified 1ad84fea

[Bugfix] handle grammar compilation failures to avoid engine crash (#47312)

izhuhaoran committed 17 days ago

Verified 12213c67

[Bugfix][Core] shm_broadcast: bound idle reader waits and release read slots (#45224)

chaeminlim-mb committed 17 days ago

Verified 10c75477

[MRV2][Spec Decode] Avoid rejection sampler OOM by chunking (#48630)

mgoin committed 17 days ago

Verified ac36a7a1

[Core] Simplify KVBlockZeroer index tensor handling (#48399)

njhill committed 17 days ago

Verified 521aa80f

[MooncakeStore] Re-derive full external hits on stored boundaries (#49481)

Dao007forever committed 17 days ago

Verified a76df87d

[Perf][KVConnector][Mooncake] Vectorize prepare_value on the KV load path (#48531)

GirasoleY committed 17 days ago

Verified a4904ba9

[CPU][Docs] Update docs and dockerfile for s390x (#49523)

R3hankhan123 committed 18 days ago

Verified f83de6d4

[Misc] Use VLLMValidationError in chat_utils content-part validation (#49217)

umut-polat committed 18 days ago

Verified 239fc735

[Bugfix] Fix DeepSeek-V4 DSpark draft shared-expert padding for TP > 8 (#49415)

mikekg committed 18 days ago

Verified 76bf5524

[Performance][Model] Avoid transient Inkling result allocations (performance, and OOM prevention on smaller memory configurations) (#49487)

mikekg committed 18 days ago

Verified 9a698f32

[Bugfix][Model] Remove SciPy dependency from Inkling scale planning (#49485)

mikekg committed 18 days ago

Verified 4080263b

[Docs] Re-add Reo.dev analytics beacon (#49474)

jcotant-inferact committed 18 days ago

Verified fc5fda10

[Bugfix] Make shared NVFP4 MoE scales writable (#49489)

S1ro1 committed 18 days ago

Verified b07ec92f

Fused Shared Expert Support for AMD Quark DeepSeek-V4 Model Checkpoints (#48044)

ColinZ22 committed 18 days ago

Verified 27ffbfde

[BugFix] Handle per-group prefix-hit divergence for hybrid models with KV connector (#48425)

njhill committed 18 days ago

Verified 229e01e9

Add quantization label automation (#49492)

mgoin committed 18 days ago

Verified 191146db

[Core][DSV4] Compact MXFP4 indexer KV cache and packed group overlays (#48993)

GirasoleY committed 18 days ago

Verified f3a920a0