[Bugfix] Fix GLM rotary_dim issue and support v1 (#16912)

Isotr0py committed over 1 year ago

[Misc] Refactor platform to get device specific stream and event (#14411)

Shanshan Shen committed over 1 year ago

[Misc] fix collect_env version parse (#15267)

wangxiyuan committed over 1 year ago

Restore buffers when wake up from level 2 sleep (#16564) (#16889)

Han Zhang committed over 1 year ago

[Doc] Split dummy_processor_inputs() in Multimodal Docs (#16915)

Alex Brooks committed over 1 year ago

[Bugfix] Fix distributed bug in Qwen2.5-VL & Qwen2.5-Omni (#16907)

Yang Fan committed over 1 year ago

[CI/CD][V1] Add spec decode tests to CI (#16900)

Woosuk Kwon committed over 1 year ago

[Bugfix] Fix v1/spec_decode/test_ngram.py (#16895)

qizixi committed over 1 year ago

[easy] Pass compile_fx only the config patches (#16845)

Richard Zou committed over 1 year ago

Improve configs - `CacheConfig` (#16835)

Harry Mellor committed over 1 year ago

Serialize tensors using int8 views (#16866)

Staszek Paśko committed over 1 year ago

Log how much time loading a compiled artifact takes (#16848)

Richard Zou committed over 1 year ago

[doc] update hyperlink (#16877)

Reid committed over 1 year ago

[VLM] Clean up models (#16873)

Cyrus Leung committed over 1 year ago

[Model] Qwen2.5-Omni Cleanup (#16872)

Roger Wang committed over 1 year ago

[Model] Refactor Phi-4-multimodal to use merged processor and support V1 (#15477)

Isotr0py committed over 1 year ago

[V1][Misc] stop update prefix cache stats when logs_stats is disabled (#16460)

vie-serendipity committed over 1 year ago

[Misc] Benchmarks for audio models (#16505)

Nicolò Lucchesi committed over 1 year ago

[Frontend] Add sampling params to `v1/audio/transcriptions` endpoint (#16591)

Nicolò Lucchesi committed over 1 year ago

[rocm][MI300] llama4 maverick fp8 moe config tp8 (#16847)

Divakar Verma committed over 1 year ago