Commits vllm-project/vllm

Fix DeepSeek V4 SM10 sparse MLA default

LucasWilkinson committed 43 days ago

79b560e2

[Bugfix][Gemma4] Fix FA4 mm_prefix mask: add sliding window and absolute q_idx (#47332)

lucianommartins committed 43 days ago

Verified 34b560b7

[Bugfix][Model] Allow Run:ai memory_limit sentinel values (#47337)

Sunt-ing committed 43 days ago

Verified 91b56473

[ROCm][CI] Fix Kernels and Kernels attention test failures (#47519)

cpersson-amd committed 43 days ago

Verified 4a6bf3c7

[Bugfix][Frontend] Preserve default sampling params in batch chat (#47597)

Sunt-ing committed 43 days ago

Verified d2afe396

[Perf] Remove redundant op for GLM 5.2 (#47198)

yewentao256 committed 43 days ago

Verified 2a9113f9

[Bugfix][Frontend][gpt-oss] Recover raw tail when Harmony parser ends non-terminal (#47379)

yzong-rh committed 43 days ago

Verified 0cd6f767

[CI] Bump `huggingface-hub` from `v1.10.2` to `v1.22.0` (#47551)

hmellor committed 43 days ago

Verified f1445f6d

[Misc] Validate Pooling cache_salt Values (#46966)

taneem-ibrahim committed 43 days ago

Verified 1d354c69

[Misc] Update request-extras parity for batch chat completion (#47333)

taneem-ibrahim committed 43 days ago

Verified 2f212245

[Misc] Forward request-level prompt extras for cross-encoder scoring (#46939)

taneem-ibrahim committed 43 days ago

Verified fa1fa968

[Misc] Preserve cross-encoder pooling extra kwargs (#47082)

taneem-ibrahim committed 43 days ago

Verified 6eac8e00

[XPU] Add W8A8 FP8 linear kernel with multi-granularity quant support (#43645)

chaojun-zhang committed 43 days ago

Verified 1a308c44

[Bugfix][Structured Output][Spec Decode] Constrain bitmask and trim grammar advance at the reasoning boundary (#44297)

yuyue0225sc committed 43 days ago

Verified e7c9df94

[Bugfix] Fix CPU split-KV scratchpad sizing (#45844)

gausah01 committed 44 days ago

Verified 26eb8720

[ROCm] Disable persistent sparse-MLA kernel for chunked-prefill continuations (#47567)

Rohan138 committed 44 days ago

Verified 4c3c17d4

[ROCm][CI][Bugfix] Use VllmRunner for `voxtral_realtime` tests to avoid OOM on AMD GPU (#47536)

shen-shanshan committed 44 days ago

Verified f329ce40

[MRV2][SD] Make Dynamic SD comatible with Full Cuda Graphs (#45953)

ekagra-ranjan committed 44 days ago

Verified 07516fda

Support nvfp4 kv with kv-cache-dtype-skip-layers sliding_window (#42890)

sychen52 committed 44 days ago

Verified 67ff0ae3

[Frontend] Limit `SO_REUSEPORT` to multi-worker serving (#47529)

BugenZhao committed 44 days ago

Verified ab3b6d97

[Frontend] [Parser] Port DeepSeek V4 to streaming parser engine framework (#45877)

bbrowning committed 44 days ago

Verified fb5291b3

[GLM4V] Avoid GLM4V processor init during startup metadata reads (#47155)

labAxiaoming committed 44 days ago

Verified d6d39c11

[Bugfix][Multimodal] Normalize direct PIL image inputs (#47566)

Sunt-ing committed 44 days ago

Verified 37995019

[AMD][EPLB] Enable EPLB for Quark OCP MXFP4 MoE (#47220)

okorzh-amd committed 44 days ago

Verified 576bf75d

[CI][AMD] Allow git operations on previously created work trees (#47554)

tpopp committed 44 days ago

Verified f006e5a2

[ROCm] Fix encoder-decoder cross-attention KV layout aliasing (#47035)

djramic committed 44 days ago

Verified f63dca68

[Rust Frontend] Speed up chat roundtrip tests (#47523)

BugenZhao committed 44 days ago

Verified 8651f043

[ROCm][CI] Adding test groups for parity with upstream (#47479)

AndreasKaratzas committed 44 days ago

Verified 3775d5fc

[CI Bugfix] Lazily import Qwen warmup dependencies (#47539)

LopezCastroRoberto committed 44 days ago

Verified d7192cfc

[Bugfix][CPU] Ship examples/ in the CPU release image (#47447)

AgenticSpark committed 44 days ago

Verified 978de833