[Bugfix]: Fix messy code when using logprobs (#19209)

Chauncey committed over 1 year ago

Make distinct `code` and `console` admonitions so readers are less likely to miss them (#20585)

Harry Mellor committed over 1 year ago

[Optimize] Don't send token ids when kv connector is not used (#20586)

Woosuk Kwon committed over 1 year ago

[Bugfix] Fix Maverick correctness by filling zero to cache space in cutlass_moe (#20167)

Ming Yang committed over 1 year ago

[Misc] Improve logging for dynamic shape cache compilation (#20573)

Kyle Yu committed over 1 year ago

[Config] Refactor mistral configs (#20570)

Patrick von Platen committed over 1 year ago

Fix links in multi-modal model contributing page (#18615)

Harry Mellor committed over 1 year ago

[Misc] feat output content in stream response (#19608)

rongfu.leng committed over 1 year ago

[Misc] Add fully interleaved support for multimodal 'string' content format (#14047)

Anton committed over 1 year ago

[Kernel] Optimize Prefill Attention in Unified Triton Attention Kernel (#20308)

jvlunteren committed over 1 year ago

[DP] Copy environment variables to Ray DPEngineCoreActors (#20344)

Rui Qiao committed over 1 year ago

[Front-end] microbatch tokenization (#19334)

ztang2370 committed over 1 year ago

[Bugfix] Prevent IndexError for cached requests when pipeline parallelism is disabled (#20486)

Peter Pan committed over 1 year ago

[Model][Last/4] Automatic conversion of CrossEncoding model (#19675)

wang.yuqi committed over 1 year ago

[Doc] Fix some MkDocs snippets used in the installation docs (#20572)

Harry Mellor committed over 1 year ago

[Doc] Add outline for content tabs (#20571)

Harry Mellor committed over 1 year ago

[Doc] Fix internal links so they don't always point to latest (#20563)

Harry Mellor committed over 1 year ago

[Doc] Use `gh-pr` and `gh-issue` everywhere we can in the docs (#20564)

Harry Mellor committed over 1 year ago

[Doc] Remove extra whitespace from CI failures doc (#20565)

Harry Mellor committed over 1 year ago

[Misc] Set the minimum openai version (#20539)

Jee Jee Li committed over 1 year ago