Use hidden_size_per_head as head_size fallback (#24221)

nopperl committed about 1 year ago

[Model] Add pp support for hunyuan (#24212)

Jiangyun Zhu committed about 1 year ago

[Doc] Update vLLM Singapore Meetup info (#24234)

TJian committed about 1 year ago

[Feature][Response API] Add streaming support for non-harmony (#23741)

Kebe committed about 1 year ago

[Hardware][Apple-CPU] Disable OneDNN build for Apple Silicon (#24200)

Ignacio Sica committed about 1 year ago

[Attention] FlashAttn MLA (#14258)

Lucas Wilkinson committed about 1 year ago

[Bugfix] Fix Incremental Detokenization with `tokenizers == 0.22.0` (#24159)

Fanli Lin committed about 1 year ago

[Attention][Platform] Refactor MLA to support Custom Op (#23332)

whx committed about 1 year ago

Improve flexibility of auto_tune.sh execution. (#23766)

anthonsu committed about 1 year ago

[Core][Model] Terratorch backend integration (#23513)

mgazz committed about 1 year ago

[Model] Add MiDashengLM model support (#23652)

bingchen-mi committed about 1 year ago

[Misc] Enhance output readability of helper script (#24214)

Weida Hong committed about 1 year ago

[CPU] Refactor CPU unquantized linear (#24150)

Li, Jiang committed about 1 year ago

Migrate ultravox inputs to TensorSchema (#23503)

Benji Beck committed about 1 year ago

[Refactor] Introduce basic Renderer for completion-style request (#24010)

Flora Feng committed about 1 year ago

[Kernel][Bugfix] Fix grouped topk cu (#24146)

Qiming Zhang committed about 1 year ago

[Feature][Responses API]Support MCP tools with streaming mode + background mode (#23927)

wuhang committed about 1 year ago

Remove deprecated `PyNcclConnector` (#24151)

Peter Pan committed about 1 year ago

[Feature][gpt-oss] Add support for num_cached_tokens and num_reasoning_tokens tracking (#23460)

George Nagy II committed about 1 year ago

[Bugfix][DP] DP distribution does not require ray[default] (#23822)

Kebe committed about 1 year ago