Move query quantization to attention layer for Flashinfer & Triton. (#26534)

Adrian Abeyta committed about 1 year ago

[Bug] Temporally Disable `VLLM_ALLREDUCE_USE_SYMM_MEM` by Default (#26925)

Wentao Ye committed about 1 year ago

support flashinfer_fp4 moe for 5090 gpu (#26669)

XiaobingZhang committed about 1 year ago

Vectorize RMS norm variance using vectorize_read_with_alignment (#26234)

Benji Beck committed about 1 year ago

[NVIDIA] Add support for cudnn fp4 gemm via flashinfer (#26107)

Kaixi Hou committed about 1 year ago

[Chore] Clean up CODEOWNERS (#26923)

Woosuk Kwon committed about 1 year ago

[ModelOpt] Remove NVFP4 MoE K%16==0 constraint (#26891)

XiaobingZhang committed about 1 year ago

[Feature]: Use pydantic validation in observability.py config (#26637)

Sam/Samuel committed about 1 year ago

Olmo 3 tool parser and tests (#26143)

Pradeep Dasigi committed about 1 year ago

Lower sevarity of log when model info cache misses due to exception (#26917)

Harry Mellor committed about 1 year ago

[Chore] Separate out `vllm.utils.async_utils` (#26913)

Cyrus Leung committed about 1 year ago

[Chore] Separate out `vllm.utils.func` (#26904)

Cyrus Leung committed about 1 year ago

[BugFix] Patch inductor memory plan logic (#26878)

Boyuan Feng committed about 1 year ago

chore: remove unused marker (#26890)

Max Wittig committed about 1 year ago

[Misc] rename torch_dtype to dtype (#26695)

wangxiyuan committed about 1 year ago

[Misc] Remove `isort` and `yapf` ignores (#26888)

Cyrus Leung committed about 1 year ago

[Model][2/N] Improve all pooling task | Support multi-vector retrieval (#25370)

wang.yuqi committed about 1 year ago

[Lora]Load tuned multi-lora kernel configs from json files (#26319)

li2haipeng committed about 1 year ago

[Platform] allow platform to init dp group (#22243)

wangxiyuan committed about 1 year ago

[Easy] Get rid of unnecessary paraenthesis in kv_cache_manager (#26842)

Jialin Ouyang committed about 1 year ago