Commits vllm-project/vllm

Update async_llm_streaming.py

khluu committed 25 days ago

Verified f500b652

[CI] fix docker image build by specifying merge-base commit id when downloading pre-compiled wheels (#29930)

Harry-Chen committed 25 days ago

Verified 1109f982

[Bugfix] Respect VLLM_CONFIGURE_LOGGING value (#28671)

elizabetht committed 25 days ago

Verified b5407869

[Kernels] Remove BatchedTritonOrDeepGemmExperts and default fallback to Triton (#29929)

bnellnm committed 25 days ago

Verified 2902c348

[CI] Fix re import error (#29973)

yewentao256 committed 25 days ago

Verified ac188658

Fix LLMEngine.del dp_group cleanup condition (#29954)

hyongtao-code committed 25 days ago

Verified 2fc5d6e0

[Bugfix] Fix flashinfer ar+norm kernel not available issue (#29960)

elvischenv committed 25 days ago

Verified afe9eb40

[Performance][DP/EP] Add silu_mul_per_token_group_quant_fp8_colmajor kernel (#29470)

varun-sundar-rabindranath committed 25 days ago

Verified 19bee6d1

[Bugfix] Mistral tool parser streaming update (#19425)

avigny committed 25 days ago

Verified dd5d1ef7

[ROCm][CI] Fix v1/logits_processors failure on ROCm (#29927)

micah-wil committed 25 days ago

Verified d1f7392c

SigLIP example add chat_template (#29902)

piood committed 25 days ago

Verified 9ae3c55b

[Core] Add xxHash as a high-performance hash option for accelerating prefix caching (#29163)

LuminolT committed 25 days ago

Verified 9bcf9229

[CI/Build][AMD] Skip test_shared_storage_connector_hashes in test_shared_storage_connector.py due to hipErrorLaunchFailure when calling .cpu() (#29839)

rasmith committed 25 days ago

Verified 5aa9b090

[CPU Backend] [Doc]: Update Installation Docs for CPUs (#29868)

ioghiban committed 25 days ago

Verified 1bb17ecb

[GPU Backend] [Doc]: Remove duplicate statements on missing GPU wheels. (#29962)

ioghiban committed 25 days ago

Verified 15b1511a

[Frontend] supports deepseekv32 chat template (#29837)

chaunceyjiang committed 25 days ago

Verified b78772c4

[docker] Build CUDA kernels in separate Docker stage for faster rebuilds (#29452)

amrmahdi committed 25 days ago

Verified f5d3d93c

[DOC] Add Arm to list of compute resouces providers (#29894)

fadara01 committed 25 days ago

Verified 78f4bb0b

[refactor] CTMoEMethods to use QuantizationArgs (#28871)

HDCharles committed 25 days ago

Verified b294e28d

[Bugfix] Follow-up fix on MediaWithBytes (#29951)

ywang96 committed 25 days ago

Verified 787b84a9

[Bugfix][Quantization] Support BF16 tensors on GGUF (#29948)

a4lg committed 25 days ago

Verified 42c19496

[CI/Build] Avoid duplicate empty inputs test for common multimodal generation tests (#29907)

Isotr0py committed 25 days ago

Verified cc4e296e

[Bugfix] Fix incorrect `image_grid_thw` rank for HunyuanOCR from missing `merge_by_field_config=True` (#29950)

Isotr0py committed 25 days ago

Verified a21cd9ed

[CI] Add Async Eplb nightly CI tests (#29385)

david6666666 committed 25 days ago

Verified 7fe9c1a2

[Refactor] [1/N] to simplify the vLLM serving architecture (#28040)

chaunceyjiang committed 25 days ago

Verified 3f42b05f

Add logging for cudagraph related info (#29825)

sarckk committed 26 days ago

Verified 69520bc6

[responsesAPI] support input output messages for non harmony models (#29549)

qandrew committed 26 days ago

Verified 3a775148

[Misc] Allow `fetch_*` utils to access local files by default (#29932)

DarkLight1337 committed 26 days ago

Verified bbfb55c2

[BugFix] fix imgs_pos in hunyuan_vl (#29879)

wkcn committed 26 days ago

Verified 0bec63fa

[Bugfix] Defunctionalize TRTLLM AR+Norm op for avoiding extra clone kernel before it (#29631)

elvischenv committed 26 days ago

Verified c719c405