vllm-project/vllm

Pull Requests Commits

TMP: no cleanup post-lowering

ProExpertProg committed 1 day ago

57e7b371

Disable fusions for batch-invariant

ProExpertProg committed 1 day ago

8ac7a8ae

Merge branch 'main' into luka/vllm-ir/rms-norm-batch-invariant

ProExpertProg committed 3 days ago

a4065fee

Update registry for Nemotron-v3 VL Nano/Super (#39747)

collinmccarthy committed 3 days ago

Verified 27c0ca50

ProExpertProg committed 3 days ago

ba691125

[CI Bug] fix flaky test (#39938)

yewentao256 committed 3 days ago

Verified 7c636432

[FlashAttention] Don't overwrite `flash_attn_interface.py` when installing precompiled (#39932)

MatthewBonanni committed 3 days ago

Verified c77e596e

[Bugfix][Perf] Indexer upcast WK to BF16 for fusion (#38928)

benchislett committed 3 days ago

Verified ac3dac54

[Bug] Fix batch invariant test issue, bs=1 with `max_seq_num = 1` (#39320)

yewentao256 committed 3 days ago

Verified 39ac6404

[Speculative Decoding] Add DFlash speculators config parsing (#38300)

ZhanqiuHu committed 3 days ago

Verified 0b790a25

[Bugfix][NIXL] Fix `_logical_to_kernel_block_ids` conversion for non-mamba models (#39724)

ZhanqiuHu committed 3 days ago

Verified 41488f2a

[CI] Only build release Docker images when NIGHTLY=1 (#39882)

khluu committed 3 days ago

Verified 102d51c9

[Mooncake] Fix mixed MLA+Eagle block-size validation (#39596)

zhewenl committed 3 days ago

Verified 55e1a8e1

Bug/test eagle dp v2 (#39838)

Monishver11 committed 4 days ago

Verified 21e5a9f4

[Test] Fix @create_new_process_for_each_test("fork") in interactive shell pipeline (#29130)

markmc committed 4 days ago

Verified 8ad6ff00

[BugFix] KeyError on scope["method"] for realtime api websocket in AuthenticationMiddleware (#36934)

daniebrill committed 4 days ago

Verified f2145efc

[KVConnector][LMCache] Propagate cache_salt through MP connector for per-user cache isolation (#39837)

royyhuang committed 4 days ago

Verified ed333105

[SpecDecode][Benchmark] Add SPEED-bench support to benchmarking CLI (#36029)

talorabr committed 4 days ago

Verified 3cc328a4

[XPU] properly handle q_descale on XPU as quant query input not supported (#39676)

yma11 committed 4 days ago

Verified 3beb57a2

FIX: support language_model.backbone naming in NemotronH Nano VL quantization config (#39901)

danielafrimi committed 4 days ago

Verified 8b553193

[BugFix][Graph] fix: handle empty sym_shape_indices in PiecewiseBackend. (#39395)

chaunceyjiang committed 4 days ago

Verified db8d4a4a

[XPU][MXFP4] add mxfp4 quant op for XPU (#39857)

zufangzhu committed 4 days ago

Verified fc701c80

[Metrics] Add request_id to FinishedRequestStats to enable correlation between metrics and requests (#39710)

Csrayz committed 4 days ago

Verified 68be0f85

[Quantization][Autoround][CPU] Add W4A16 Support (#38192)

Zhenzhong1 committed 4 days ago

Verified 60995c05

fix online fp8 for MiniCPM models (#39862)

yma11 committed 4 days ago

Verified 29e5d102

[kv_offload+HMA][3/N]: Remove block_size from KVEvents (#36644)

orozery committed 4 days ago

Verified 235e1f93

[Bugfix] Fix tool_calls Iterable consumed when debug logging is enabled (#34844)

wojciech-wais committed 4 days ago

Verified 431cea3e

[CI][NIXL] Fix PD CI breakage: pin nixl-cu{12,13} versions (#39851)

ZhanqiuHu committed 4 days ago

Verified 799973af

[Bugfix] Respect VLLM_WEIGHT_OFFLOADING_DISABLE_PIN_MEMORY in prefetch offloader (#37699)

he-yufeng committed 4 days ago

Verified bcc2306c

[Test] Refactor hard coded device string in test files under compile/quantization/models/model_executor folders (#38901)

wincent8 committed 4 days ago

Verified 3abf8584

Older