vllm-project/vllm

Pull Requests Commits

perf(deepseek): Multi-stream overlap of indexer wk+weights_proj with QKV-A for DeepSeek V3.2 NSA

alexm-redhat committed 28 days ago

cefb9334

[Misc] Move `pyav` and `soundfile` to common requirements (#39997)

Isotr0py committed 28 days ago

Verified 617d1c2f

[Bugfix] Fix Ray compiled-DAG SHM channel stalls by detaching zero-copy `np.ndarray` logprobs buffers (#35736)

JeanPaulShapo committed 28 days ago

Verified 692db29c

[Refactor] Remove `resampy` dependency (#39524)

Isotr0py committed 28 days ago

Verified 82531edb

[Misc] `toy_proxy_server` handle min_tokens (#39706)

NickLucche committed 28 days ago

Verified 3daca38e

[Bugfix] Fix LLM priority normalization for single-string prompts (#40011)

daiyu1111 committed 28 days ago

Verified a302a8fd

[Frontend][last/5] Improve pooling entrypoints | clean up. (#39675)

noooop committed 28 days ago

Verified 4e8c3f1c

[Doc] add docs for online quant frontend (#39736)

vkuzo committed 28 days ago

Verified 5e5afafa

[CI/Build] Improve stability of CPU tests (#39966)

bigPYJ1151 committed 28 days ago

Verified 324a3d2b

[Model] Use mm_features to compute mrope positions for PaddleOCR-VL (#39888)

grYe99 committed 28 days ago

Verified 4269b794

[Kernel][Helion] Fix inductor fusion of Helion HOP (#39944)

gmagogsfm committed 28 days ago

Verified edc36489

[Nixl] Bump Nixl version to 0.10.1 (#39922)

NickLucche committed 28 days ago

Verified 9965f501

[Model] Use mm_features for Keye-VL and Keye-1.5-VL M-RoPE (#39869)

lalit10 committed 28 days ago

Verified 17d87168

Fix #33773: Replace unconditional pandas import with PlaceholderModule (#39990)

netanel-haber committed 28 days ago

Verified 98700c61

[Docs] Update PR template to remove release notes google docs (#39982)

simon-mo committed 28 days ago

Verified 10e49d26

[Bugfix] Accept **kwargs in MiniMaxM2Parser.__init__() (#39861)

SeraphimSerapis committed 28 days ago

Verified 8d7c9628

[XPU] use spawn multiproc method on xpu (#39671)

xinyu-intel committed 28 days ago

Verified f4ddaf8c

Add Jina Embeddings v5 model support (fixes #38633) (#39575)

Roy214 committed 28 days ago

Verified 2cdf8604

[Bugfix] add support for 'num_attention_groups' in ModelArchConfigConvertorBase for Step3p5 (#39796)

realliujiaxu committed 28 days ago

Verified 78453792

[CPU][IBM Z][Dockefile][Docs] Fix s390x builds for torch 2.11 and update docs for s390x (#39910)

R3hankhan123 committed 28 days ago

Verified 4b7ca37b

[perf][cpu] Accelerate BF16 GELU with LUT impl on Arm CPUs (#37469)

fadara01 committed 28 days ago

Verified 445b7093

[Bugfix] Reject empty tools array with HTTP 400 (#39780)

jigangz committed 28 days ago

Verified 18013df6

[Mistral Grammar] Fix tool and reasoning parsing (#39217)

juliendenize committed 28 days ago

Verified c0722f22

[compile] Invoke split FX graph by codegen. (#38657)

zhxchen17 committed 28 days ago

Verified 951dca80

[ROCm][FEAT] Integrate aiter gemm w8a8 ptpc (#33773)

vllmellm committed 28 days ago

Verified 5f7fab88

[Model Runner V2][BugFix] fix num_sampled dtype for probabilistic rej… (#39951)

TheEpicDolphin committed 29 days ago

Verified 343f6523

[Quantization] - Layerwise reloading of Attention/KV quantized models (#38995)

Josephasafg committed 29 days ago

Verified 19fa90ed

Update to transformers v5 (#30566)

hmellor committed 29 days ago

Verified 03f8d3a5

[Model] Fix Gemma 4 token repetition by dynamic BOS injection for PT models (#39842)

lucianommartins committed 29 days ago

Verified 6dc94914

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

collinmccarthy committed 29 days ago

Verified 27c0ca50

Older