[Bugfix] fix RAY_CGRAPH_get_timeout is not set successfully (#19725)

Chauncey committed over 1 year ago

[Hardware][AMD] integrate aiter chunked prefill into vllm (#18596)

Zzz9990 committed over 1 year ago

[Qwen] Add tagging rule for Qwen related PRs (#19799)

Lu Fang committed over 1 year ago

[Platform] Allow platform use V1 Engine by default (#19792)

wangxiyuan committed over 1 year ago

[doc] fix the incorrect label (#19787)

Reid committed over 1 year ago

[Minor] Zero-initialize attn output buffer (#19784)

Woosuk Kwon committed over 1 year ago

[V1] Decouple GPU and TPU `InputBatch` (#19778)

afeldman-nm committed over 1 year ago

[V1][P/D] An native implementation of xPyD based on P2P NCCL (#18242)

Zhonghua Deng committed over 1 year ago

[V1] Add API docs for EncoderCacheManager (#19294)

Russell Bryant committed over 1 year ago

[Misc] Add __str__ for RequestStatus (#19780)

lkchen committed over 1 year ago

[MISC] correct DeviceConfig device field static type analysis (#19699)

Ning Xie committed over 1 year ago

[MISC] correct copy_blocks src_to_dists param type (#19696)

Ning Xie committed over 1 year ago

[TPU] Update torch version to include paged attention kernel change (#19706)

Chenyaaang committed over 1 year ago

[Feature][ROCm] Add full graph capture support for TritonAttentionBackend (#19158)

Charlie Fu committed over 1 year ago

[Bugfix] Fix faulty triton importing logic when using Ray for DP (#19734)

Michael Goin committed over 1 year ago

[Misc] Update lmcache connector with the latest connector apis (#19441)

Jiayi Yao committed over 1 year ago

Remove sm120 arch from sm100 cutlass kernel arch list (#19716)

Michael Goin committed over 1 year ago

[Perf] Optimize `moe_align_block_size` CUDA kernel (#19572)

Wentao Ye committed over 1 year ago

[Bugfix] Update multimodel models mapping to fit new checkpoint after Transformers v4.52 (#19151)

Isotr0py committed over 1 year ago

[Mis] remove duplicate engine status checks (#19647)

CYJiang committed over 1 year ago