[Bugfix][NIXL] Fix DCP pulls across MLA cache regions (#57389)

Lucas Wilkinson committed about 1 month ago

[docs] Fix legacy hf CLI references (vllm) (#57958)

Lucain committed about 1 month ago

[Bugfix] Fix external LB DP rank handling when replicas share nodes (#53743)

Tony Lin committed about 1 month ago

[ROCm][DSv4] Enable DSpark adaptive verification (#52362)

Tuukka Sarvi committed about 1 month ago

[Bugfix][Frontend] Accept diarized transcription responses in run-batch (#57948)

sfigs committed about 1 month ago

[Pooling] Fix normalization of chunked long-text embeddings (#57498)

Taneem Ibrahim committed about 1 month ago

[MM] Move `supports_multimodal_inputs` and cache out of registry (#57913)

Cyrus Leung committed about 1 month ago

[ROCm][Tests] Reduce host memory when downcasting FP32 HF references (#57869)

Andreas Karatzas committed about 1 month ago

[Frontend] Add streaming parity tests and docs for derender (#57922)

Martin Hickey committed about 1 month ago

[Bugfix][Multimodal] Tolerate malformed EXIF in Molmo 2 image preprocessing (#57234)

Hotragn Pettugani committed about 1 month ago

[CI][XPU] Deselect Ray UT in XPU V1 test Job (#57889)

Chaojun Zhang committed about 1 month ago

[Security] Prefer fresh multimodal payloads over a stale receiver cache (#57833)

Juan Pérez de Algaba committed about 1 month ago

[CI][ROCm] Add ten AMD parity groups (#57876)

Andreas Karatzas committed about 1 month ago

[Test] Organize structured output utility tests (#57545)

CaiJohn committed about 1 month ago

[Bugfix][ROCm][DSv4.1] Disable SWA bounded replay on ROCm (#57906)

Fangzhou Ai committed about 1 month ago

[AuxOutput] Add block-keyed storage for routed-expert outputs (#45635)

Hongxin Xu committed about 1 month ago

[CI][ROCm] Increase timeout for MI300 Distributed DP Extended (#57903)

djramic committed about 1 month ago

[CI][ROCm] Mirror remaining portable suites without duplicate coverage (#57877)

Andreas Karatzas committed about 1 month ago

[AMD][Minimax-M3][perf] Enable packed LBHNC AITER QK-norm fusion for MiniMax-M3 on ROCm (#54535)

weitliao committed about 1 month ago

[CI][ROCm] Make Python-only installation failures blocking (#57870)

Andreas Karatzas committed about 1 month ago