Commits vllm-project/vllm

Reduce SP correctness test matrix from 32 to 4 cases

tlrmchlsmth committed 126 days ago

Verified 5ad0151d

factor out _build_anthropic_usage helper

tlrmchlsmth committed 126 days ago

Verified 8019b6ec

Merge branch 'main' into anthropic-cache-usage

zhangshuoming990105 committed 128 days ago

eb3ee8af

Merge fork branch updates

zhangshuoming990105 committed 128 days ago

77648ab2

Address review on Anthropic cache usage reporting

zhangshuoming990105 committed 128 days ago

fd3e0cac

Merge branch 'main' into anthropic-cache-usage

zhangshuoming990105 committed 128 days ago

82d1ddf3

[Bugfix] Fall back to Pydantic loc for param in validation errors (#46038)

muhammadfawaz1 committed 128 days ago

Verified 69bdd345

[xpu] bump up vllm-xpu-kernels v0.1.10 and upgrade 2618 umd (#40367)

jikunshang committed 128 days ago

Verified ec67d7ae

Merge branch 'main' into anthropic-cache-usage

zhangshuoming990105 committed 128 days ago

7341ff15

[AMD][CI] Fix Language Models Test (Extended Generation) failures (#45509)

okorzh-amd committed 128 days ago

Verified ecf9d835

[Docs] Update stale LMCache examples (#45762)

sammshen committed 128 days ago

Verified c9135db2

[DeepSeek-V4] Support TEP=16 for the block-FP8 shared expert (#46001)

majunze2001 committed 128 days ago

Verified 2a6c6b94

[bugfix]Indexer init skip and MTP TopK share for iteration (#45895)

JaredforReal committed 128 days ago

Verified ab666069

[Bugfix] Fix corrupt outputs in MoE FP8 LoRA responses and MoE base model responses when LoRAs are loaded (#42120)

nv-nedelman-1 committed 128 days ago

Verified 9ea3a401

[Cohere] Remove dead prepare_structured_tag override in Cohere parser (#46099)

sfeng33 committed 128 days ago

Verified 560fb8b8

[Model Runner V2] Fix MRv2 memory leak test (#46095)

yewentao256 committed 128 days ago

Verified 675cd5d2

[Bugfix] [Parser] Fix empty tool block silently dropping subsequent content (#46091)

bbrowning committed 128 days ago

Verified 7f616c32

[Perf] Remove unused loggers in `reasoning/` (#45988)

IvyXu420 committed 128 days ago

Verified c3c6d723

[Bugfix][KV Connector] Disable Mooncake TP put-striding when DCP > 1 (#45371)

ivanium committed 128 days ago

Verified 41dcf49c

[KV Connector][Mooncake] Async lookup to reduce scheduler overhead (#45659)

ivanium committed 128 days ago

Verified 35e4dd4a

fix(anthropic): auto-detect template support for mid-conversation system messages (#46025)

felix0080 committed 128 days ago

Verified 4ce2d014

[MRV2] Make FP32 Gumbel sampling more accurate (#45996)

WoosukKwon committed 128 days ago

Verified 16908e13

[CI Bug] Revert #42379 to fix CI `Multi-Modal Models (Extended Generation 1)` (#46070)

yewentao256 committed 128 days ago

Verified 225936a1

(security) Upgrade Starlette to >= 1.0.1 to fix CVE-2026-48710 (#45675)

jperezdealgaba committed 128 days ago

Verified f6ba7209

[Model Runner V2] Migration to support quantized model by default [5/N] (#44446)

yewentao256 committed 128 days ago

Verified b53b1c7f

[Bugfix][Quantization] Don't reject fp8_e5m2 KV cache for non-fp8 quantized checkpoints (#45040)

Sunt-ing committed 128 days ago

Verified 79ca54d2

[Bugfix] [Parser] Fix Qwen3 latent bug in partial params dropping values containing `<` (#46047)

bbrowning committed 128 days ago

Verified 09f3cd5c

[KV Connector][Offloading] Disable parallel-agnostic fs-tier cache on V2 model runner (#46044)

Etelis committed 128 days ago

Verified ea6078fe

[Tests] Add Qwen3 streaming parser delta boundary cases (#45708)

Palaiologos1453 committed 128 days ago

Verified a0df04e4

[ROCm][Spec Decode] Fix probabilistic draft probs test attention backend (#45706)

stefankoncarevic committed 128 days ago

Verified e2352c29