Fix V1 engine serialization error with Ray distributed executor (#26148)

Nikhil G committed about 1 year ago

Avoid division by zero in cache DS MLA kernel (#26174)

Matthew Bonanni committed about 1 year ago

Stop mergify from keeping stale PRs alive (#26169)

Harry Mellor committed about 1 year ago

[BugFix][QWEN-VL]fix wrong apply_rotary_emb_torch selection introduced by #24642 (#26123)

Chendi.Xue committed about 1 year ago

Fix undefined symbol: cutlass_moe_mm_sm100 (#26098)

Jun Jiang committed about 1 year ago

[Renderer] Move Processor out of LLMEngine (#26165)

Cyrus Leung committed about 1 year ago

[Model] Fixed stream generator for gpt-oss + spec-decoding (#26027)

Aleksandr Samarin committed about 1 year ago

[CI/Build] do not enforce precompilation on tpu ci tests (#25992)

Xiang Si committed about 1 year ago

[Model] Supplement to PR 24862: Pass param prefix to LLMHead (#25805)

whx committed about 1 year ago

[backends][short_conv] CUDA graph piecewise edits (#24215)

Paul Pak committed about 1 year ago

[Bugfix] Re-enable prefill of max model length (#24446)

Yannick Schnider committed about 1 year ago

[Renderer] Move Processor out of AsyncLLM (#24138)

Yang Liu committed about 1 year ago

Quick fix for IMA with the Prefix Prefill kernel during graph capture (#25983)

Sage Moore committed about 1 year ago

[Doc] Fixed shape description for fused_batched_moe.py (#25668)

Egor committed about 1 year ago

[Multi Modal] Configurable MM Profiling (#25631)

Wenlong Wang committed about 1 year ago

[openai] Fix missing tool usage check (system message) (#24768)

kyt committed about 1 year ago

[NIXL][Misc] Expose metrics from NIXL for logging to CLI (#25388)

Nicolò Lucchesi committed about 1 year ago

[CI] Fix distributed hybrid tests in CI (#26155)

Thomas Parnell committed about 1 year ago

[test utils] correct wrong typing (#26159)

Yannick Schnider committed about 1 year ago

[Model] Use `merge_by_field_config` for MM models (InternVL family) (#26153)

Cyrus Leung committed about 1 year ago