vllm-project/vllm

Pull Requests Commits

ywang96 committed 3 days ago

257c0c5b

ywang96 committed 3 days ago

e9855c5c

ywang96 committed 4 days ago

7cd88244

Fix tool_parser_cls type annotation from Callable to type[ToolParser] (#37957)

sfeng33 committed 5 days ago

Verified 2e67fa75

[KV Offload] Refactor CPU offloading: pluggable CachePolicy, remove Backend abstraction, restructure into `cpu/` package (#37874)

ronensc committed 5 days ago

Verified e3c6c10c

[Frontend][Bugfix] Pass default_chat_template_kwargs to AnthropicServingMessages (#37899)

jetxa committed 5 days ago

Verified 16a664df

[release] Move agent queue to Release cluster queues (#37783)

khluu committed 5 days ago

Verified 7281199a

Downsize CPU jobs to use small queue (#37913)

khluu committed 5 days ago

Verified b2dd75eb

[V0 Deprecation] Refactor kv cache from list to element (#37487)

yewentao256 committed 5 days ago

Verified c59a132f

[ROCm][CI] Split Entrypoints Integration (API Server 1) into 3 jobs (#37906)

AndreasKaratzas committed 5 days ago

Verified de99d91e

[CI] Add batch invariant test: Block FP8 + small MOE (#37895)

yewentao256 committed 5 days ago

Verified 83c9d525

[Model Runner V2] Gather multimodal embeddings before draft model postprocess (#37932)

TheEpicDolphin committed 5 days ago

Verified 8f4824b6

[Test] E2E Nemotron-3-Super tests (#36803)

roikoren755 committed 5 days ago

Verified 56777b5c

[CI] Split V1 Others into 3 separate jobs (#37016)

khluu committed 5 days ago

Verified 2488a82f

[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (#35007)

WindChimeRan committed 5 days ago

Verified dc6908ac

[Bug][MoE] Strengthen _supports_current_device() checks in the TRTLLM FP8, NVFP4, and FlashInfer CuteDSL MoE experts (#36728)

yzong-rh committed 5 days ago

Verified e85f8f09

[Bug][MoE] Fix TRTLLM NVFP4 Routing Kernel Precision (#36725)

robertgshaw2-redhat committed 5 days ago

Verified 5bf3c42d

[Sparse24] [Deprecation] Remove Sparse24 CT integration and kernels (#36799)

kylesayrs committed 5 days ago

Verified 38364a7e

[Async][Spec Decoding] Zero-bubble async scheduling + spec decoding (#32951)

MatthewBonanni committed 5 days ago

Verified fafe76b4

[MRV2] Consider spec decoding in warmup (#37812)

WoosukKwon committed 6 days ago

Verified ffb5b32b

[CI] split Entrypoints Integration (API Server 1) into 3 jobs (#37882)

jikunshang committed 6 days ago

Verified 91fd695b

[CI][PD] Add Hybrid SSM integration tests to CI (#37657)

NickLucche committed 6 days ago

Verified 1cbbcfe8

Use lazy graph module during split_module to defer recompile() (#37609)

angelayi committed 6 days ago

Verified aceadb5e

[Bugfix] Fix RoBERTa position_ids accumulation on CUDA graph padding (#37884)

he-yufeng committed 6 days ago

Verified ec228061

[Bugfix] RoBERTa position_id accumulation in CUDA graph padding region (#37873)

yanghui1-arch committed 6 days ago

Verified 7151ae65

[Mypy] Fix mypy for `vllm/config` (#37808)

yewentao256 committed 6 days ago

Verified 45bd5c8e

[ROCm] fix sleep mode not releasing GPU memory problem on ROCm (#37533)

aaab8b committed 6 days ago

Verified 10a1018c

[Bugfix][LoRA] Fix incorrect LoRA Log (#37877)

jeejeelee committed 6 days ago

Verified aec2dc6c

[Bugfix] Fix CPU backend crash in KV cache block zeroing (#37550)

DorBernsohn committed 6 days ago

Verified 7938d121

[XPU][MoE Refactor] Refactor xpu mxfp4 support into oracle (#37784)

jikunshang committed 6 days ago

Verified debd6e76

Older