vllm-project/vllm

Pull Requests Commits

[GDN] Use CPU tensors to build GDN metadata (#34498)

WoosukKwon committed 253 days ago

Verified 0916e796

[Feature] Pipeline Parallel Async send/recv, 2.9% E2E throughput improvement (#33368)

yewentao256 committed 253 days ago

Verified 3d2a026f

[Core] Move pause and resume functions into engine (#34125)

hao-aaron committed 253 days ago

Verified dddbff46

[KVConnector] Clean up redundant code in KV connectors (#34147)

hickeyma committed 253 days ago

Verified 47e9b63e

[Perf] fused_moe: add int4_w4a16 benchmark support and tuning config (#34130)

mgehre-amd committed 253 days ago

Verified 934acdde

[Bugfix] fix the import path in moe test utils.py (#34245)

michalowski-arm committed 253 days ago

Verified 742d214d

[Bug Fix] Fix MambaManager.cache_blocks() crash on null blocks in align mode (#34418)

haosdent committed 253 days ago

Verified 4137c5df

[BugFix] Fix and optimize max_num_blocks_per_req calculation for MambaSpec (#34440)

peakcrosser7 committed 253 days ago

Verified 7a8a46dd

[New Model] support new model ovis2.6 (#34426)

myselvess committed 253 days ago

Verified bcf0731a

[Refactor] Call renderer for online IO processor request (#34490)

DarkLight1337 committed 253 days ago

Verified ec090c24

[Bugfix] Fix mamba state dtype setting for Qwen3-Next and Qwen3.5 (#34489)

ywang96 committed 253 days ago

Verified eea3024f

[Refactor] Pass full VllmConfig to Renderer (#34485)

DarkLight1337 committed 253 days ago

Verified 2f308214

[CI/Build] Fix CUDA re-initialization error in distributed model tests (#34491)

DarkLight1337 committed 253 days ago

Verified 1b4e8e53

[Bugfix] Fix encoder cache underestimation for GLM-4V/GLM-OCR single image (#34483)

haosdent committed 253 days ago

Verified dcf6ee85

[Bugfix] Standardize getting number of image patches/tokens (#34358)

DarkLight1337 committed 253 days ago

Verified 372b2e76

[ROCm][CI] Fix serving tokens test failures (#34047)

AndreasKaratzas committed 253 days ago

Verified 6afa587d

Add new sections to CODEOWNERS (#34309)

DarkLight1337 committed 253 days ago

Verified 94ed6cf6

[Hybrid] Fix and optimize block-aligned splitting in mamba cache align mode (#33706)

peakcrosser7 committed 253 days ago

Verified bf37812c

[Bugfix] Fix Random Dataset Prefix Length Inaccuracy (#33907)

frankwang28 committed 253 days ago

Verified b86bf441

[Kernel] [Helion] [5/N] Add Helion Autotuning infrastructure (#34025)

gmagogsfm committed 253 days ago

Verified de13dd78

[Bugfix] Delete unused redundant code in Kimi-K2.5 (#34427)

LoganJane committed 253 days ago

Verified 62788f99

[Refactor] Simplify BOS/EOS token handling (#34435)

DarkLight1337 committed 253 days ago

Verified ea5ff3a1

[Bugfix] Remove assert that's no longer valid (#34443)

bnellnm committed 253 days ago

Verified 04ea31ba

[BugFix] Add block_size validation for mamba cache align mode (#34445)

peakcrosser7 committed 253 days ago

Verified 6f019e6e

Fix num_logprobs parameter description in sampler.py (#34451)

zhuohan123 committed 253 days ago

Verified d707678d

[CI/Build] Update video URLs for testing (#34446)

DarkLight1337 committed 253 days ago

Verified fc22cae4

[Kernel] [Helion] [4/N] Add silu_mul_fp8 Helion kernel (#33373)

gmagogsfm committed 253 days ago

Verified 96161fe9

[Core] Profiler improvements and lazy initialization (#33198)

jaewonlee-fb committed 254 days ago

Verified 4453ba8d

[Core] Add sleep level 0 mode with enqueue/wait pattern (#33195)

jaewonlee-fb committed 254 days ago

Verified aa181c92

[Frontend] Enable generic structured_outputs for responses API (#33709)

alecsolder committed 254 days ago

Verified be7370da

Newer Older