vllm-project/vllm

Pull Requests Commits

Merge branch 'main' into wentao-mrv2-quantized

yewentao256 committed 2 days ago

Verified d4741973

[Kernel][Test] Extend lightning_attn and awq_triton kernel tests to XPU (#43307)

adobrzyn committed 2 days ago

Verified a947f7a4

[Bugfix] Fix test_cutlass_moe.py (#44380)

bnellnm committed 2 days ago

Verified 439203d3

[Misc] Add unit tests for pooler head classes (#44471)

taneem-ibrahim committed 2 days ago

Verified 8d9536a7

[DOC] Add INT8 W4A8 docs and Arm's supported quantization schemes (#34894)

fadara01 committed 2 days ago

Verified 3da29aa4

[ROCm][CI] Add test for Aiter unified attn kernel (#44436)

divakar-amd committed 2 days ago

Verified 06f94633

[Bugfix] Reject non-positive values for ParallelConfig int knobs (#44057)

jwzheng96 committed 2 days ago

Verified 99ef6529

[Core] Freeze garbage collector in workers after model initialization (#44363)

tlrmchlsmth committed 2 days ago

Verified 4cc78c9d

[Bugfix] MiniCPM-V-4.6 video inference crash: placeholder count mismatches visual embedding count (#44509)

tc-mb committed 2 days ago

Verified 3dbb4e0a

Add model support for granite speech plus (#43519)

zvik committed 2 days ago

Verified b21443e2

[Quant] Support compressed-tensors WNA8O8Int linears and WNInt embeddings (#44340)

mgoin committed 2 days ago

Verified 06ee2d84

[DSv4] Adding TRTLLM gen attention kernel (#43827)

zyongye committed 2 days ago

Verified b5235fca

[ROCm][CI] Specifying time outs for the lm eval models (#44255)

AndreasKaratzas committed 2 days ago

Verified 3e770367

[ROCm][CI] Stabilize memory-release in the Hybrid model generation tests (#44046)

AndreasKaratzas committed 2 days ago

Verified 6f68ca3e

[ROCm] Bump fastsafetensors to v0.3.2 from PyPI, remove git source build (#43625)

wjabbour committed 2 days ago

Verified 0c96dd64

[PD][Nixl] Mamba prefix caching mode support (#42554)

NickLucche committed 2 days ago

Verified 68f5e565

[Bugfix][Compile] Guard per_token_group_fp8_quant lookup on non-CUDA platforms (#44476)

QiliangCui2023 committed 2 days ago

Verified 9354fb1b

Add GH token to docs build pre run check (#44534)

hmellor committed 2 days ago

Verified f35b5572

Refactor CT NVFP4 linear to use a single class (#42443)

dsikka committed 2 days ago

Verified e68988a2

[Bugfix] fix EVS for qwen3-vl (#44205)

garrygale committed 2 days ago

Verified 4b87b3e8

[Attention] Mamba attention module refactor - LINEAR (#43556)

wangxiyuan committed 2 days ago

Verified 90619351

[perf] Add gemma RMS AR fusion (#42646)

jiahanc committed 2 days ago

Verified d0975a4b

Fix Kimi-K2.5 FlashInfer ViT metadata (#44493)

Kevin-XiongC committed 2 days ago

Verified 1bdc60ed

[Prefix Caching] DeepSeekv4 - Support selective prefix-cache retention for sliding-window KV cache (#43447)

wzhao18 committed 2 days ago

Verified a6183563

[CI] Reverted gitignore changes (#44497)

AndreasKaratzas committed 2 days ago

Verified 22c2e875

[Frontend] Consolidate online serving utils. (#44479)

noooop committed 2 days ago

Verified d01d0b46

[Inductor] Fast-path Inductor fallback for vllm::*/vllm_aiter::* custom ops (#42129)

okorzh-amd committed 2 days ago

Verified b4b4aaa7

[CI] Resolve release V2 docker build after ROCm CI wheels change (#44463)

AndreasKaratzas committed 2 days ago

Verified 5e2af288

[EPLB] Nixl communicator optimization. Zero-copy transfers (#41633)

ilmarkov committed 2 days ago

Verified 4f423bd5

optimize the compressor 128 split cutedsl kernel (#44230)

Jie-Fang committed 2 days ago

Verified f0cd590d

Older