Convert `.buildkite` to `ruff format` (#17656)

Harry Mellor committed over 1 year ago

[Benchmarks] Refactor run_structured_output_benchmarks.sh (#17722)

Russell Bryant committed over 1 year ago

[Bugfix] Fix marlin moe fallback logic for llama4 (#18042)

Michael Goin committed over 1 year ago

[BugFix] Set default random seed to 0 for V1 (#17929)

Woosuk Kwon committed over 1 year ago

[Bugfix][V1] Only get input embeddings w/ multi-modal models if first PP (#17916)

Jin Huang committed over 1 year ago

[v1][KVCacheManager] Avoid full cache hit by controlling max_length (#17999)

Chen Zhang committed over 1 year ago

Fix Broken macro for cutlass moe (#18049)

Driss Guessous committed over 1 year ago

[BugFix] Fix 4-GPU RLHF tests (#18007)

Nick Hill committed over 1 year ago

cleanup invalid prints (#18050)

Calvin Chen committed over 1 year ago

[Feature][V1] Support `tool_choice: required` when using Xgrammar as the `StructuredOutputBackend`. (#17845)

Chauncey committed over 1 year ago

[Bugfix] Avoid repeatedly creating dummy data during engine startup (#17935)

Cyrus Leung committed over 1 year ago

[Bugfix] Fixes for new marlin moe usage (#18017)

Michael Goin committed over 1 year ago

[ROCm]: Fix build from source failure with gcc14 and ROCm 6.3 (#13779)

Arjun Kathuria committed over 1 year ago

[Misc] Slight spelling modification (#18039)

Jee Jee Li committed over 1 year ago

Implements dual-chunk-flash-attn backend for dual chunk attention with sparse attention support (#11844)

Tao He committed over 1 year ago

[ROCm] Skip tests for quantizations incompatible with ROCm (#17905)

hissu-hyvarinen committed over 1 year ago

Remove noisy warnings from `SchedulerConfig` (#17995)

Harry Mellor committed over 1 year ago

Use NVFP4 Marlin for CompressedTensorsW4A16Fp4 (#18000)

Michael Goin committed over 1 year ago

Update some more deprecated type hinting (#17998)

Harry Mellor committed over 1 year ago

[Model] Support MiMo-7B inference with MTP (#17433)

bwshen-mi committed over 1 year ago