vllm-project/vllm

Pull Requests Commits

jeejeelee committed 10 days ago

673b186e

jeejeelee committed 10 days ago

cf052836

[BUG] Fix rank calculation in NCCLWeightTransferEngine (#36940)

hao-aaron committed 10 days ago

Verified 5e1a373d

build: update smg-grpc-servicer to use vllm extra (#36938)

slin1237 committed 11 days ago

Verified 572c776b

[Bugfix] ep_scatter kernel store-load race condition (#34991)

ivanium committed 11 days ago

Verified 55d8073d

[Model Runner V2] Some code simplification (#36929)

njhill committed 11 days ago

Verified cd32d6f5

[MoE] Add routing simulation override for MXFP4 quantized MoE (#33595)

jaewonlee-fb committed 11 days ago

Verified aaa3092f

[Speculative Decoding] Add `norm_before_fc` for gpt-oss draft models (#36545)

shubhra committed 11 days ago

Verified 87985077

[AMD][Build] Add DeepEP to ROCm Dockerfile (#36086)

rjrock committed 11 days ago

Verified a79c1c2c

[ROCm][CI] Preparing gfx90a mirroring (#36210)

AndreasKaratzas committed 11 days ago

Verified cc8f1f47

Revise environment setup in AGENTS.md (#36909)

mgoin committed 11 days ago

Verified 05b9e8ab

[Feature]: Remove Chunking From FusedMoE (#34086)

SouthWest7 committed 11 days ago

Verified 2cdf9222

[bnb] Skip moe + bnb test (#36896)

SunMarc committed 11 days ago

Verified c973ecde

Add `AGENTS.md` (#36877)

hmellor committed 11 days ago

Verified e39257a5

Update Flashinfer to 0.6.6 (#36768)

dbari committed 11 days ago

Verified cc16b24b

[Bugfix] Fix KeyError in parse_response_input for reasoning items with optional content (#34499)

jeonsworld committed 11 days ago

Verified bdc23434

[Attention] Use FA4 for MLA prefill (#34732)

MatthewBonanni committed 11 days ago

Verified f444c05c

[Bugfix] fix main branch pre-commit error (1 line change) (#36897)

SoluMilken committed 11 days ago

Verified 85199f96

[Kernel] Add FP8 KV cache support to Triton MLA decode attention (#34597)

grimulkan committed 11 days ago

Verified a1257fd1

[CI] Fix mypy pre-commit errors on main (#36882)

tdoublep committed 11 days ago

Verified abcffbba

[Hardware] Replace torch.cuda.device_count/current_device/set_device API (#36145)

jikunshang committed 11 days ago

Verified 53ec16a7

[Perf] Add TRTLLM FP8 MoE Modular Kernel (#36307)

wzhao18 committed 11 days ago

Verified 2e693f48

[CI] Fix mypy for vllm/reasoning (#35742)

hickeyma committed 11 days ago

Verified 7f1f36bf

[docs] Add lightweight AI assisted contribution policy (#30947)

markmc committed 11 days ago

Verified 5282c7d4

[Perf] add packed recurrent fast path for decode (#36596)

caozuoba committed 11 days ago

Verified 9e19f833

[Frontend] Split `OpenAIServingModels` into `OpenAIModelRegistry` + `OpenAIServingModels` (#36536)

sagearc committed 11 days ago

Verified 06e0bc21

[Bugfix] Fix crash when tool_choice=required exceeds max_tokens (#36841)

chaunceyjiang committed 11 days ago

Verified 5a71cdd7

[MM][OOT] Support CPU `seq_lens` for OOT MMEncoderAttention kernels (#36605)

shen-shanshan committed 11 days ago

Verified f0d3658c

[UX] Only show FP4 Marlin fallback warning for w4a4 models (#36806)

mgoin committed 11 days ago

Verified 57431d82

[Bugfix] Warm up Triton autotuner for GDN layers during V1 profiling (#36599)

AuYang261 committed 11 days ago

Verified 3e64fe4a

Older