[XPU] Add W8A8 FP8 linear kernel with multi-granularity quant support (#43645)

Chaojun Zhang committed 4 months ago

[Bugfix][Structured Output][Spec Decode] Constrain bitmask and trim grammar advance at the reasoning boundary (#44297)

yue.yu committed 4 months ago

[Bugfix] Fix CPU split-KV scratchpad sizing (#45844)

gausah01 committed 4 months ago

[ROCm] Disable persistent sparse-MLA kernel for chunked-prefill continuations (#47567)

Rohan Potdar committed 4 months ago

[ROCm][CI][Bugfix] Use VllmRunner for `voxtral_realtime` tests to avoid OOM on AMD GPU (#47536)

Shanshan Shen committed 4 months ago

[MRV2][SD] Make Dynamic SD comatible with Full Cuda Graphs (#45953)

Ekagra Ranjan committed 4 months ago

Support nvfp4 kv with kv-cache-dtype-skip-layers sliding_window (#42890)

sychen52 committed 4 months ago

[Frontend] Limit `SO_REUSEPORT` to multi-worker serving (#47529)

Bugen Zhao committed 4 months ago

[Frontend] [Parser] Port DeepSeek V4 to streaming parser engine framework (#45877)

Ben Browning committed 4 months ago

[GLM4V] Avoid GLM4V processor init during startup metadata reads (#47155)

labAxiaoming committed 4 months ago

[Bugfix][Multimodal] Normalize direct PIL image inputs (#47566)

Ting SUN committed 4 months ago

[AMD][EPLB] Enable EPLB for Quark OCP MXFP4 MoE (#47220)

Oxana Korzh committed 4 months ago

[CI][AMD] Allow git operations on previously created work trees (#47554)

Tres committed 4 months ago

[ROCm] Fix encoder-decoder cross-attention KV layout aliasing (#47035)

djramic committed 4 months ago

[Rust Frontend] Speed up chat roundtrip tests (#47523)

Bugen Zhao committed 4 months ago

[ROCm][CI] Adding test groups for parity with upstream (#47479)

Andreas Karatzas committed 4 months ago

[CI Bugfix] Lazily import Qwen warmup dependencies (#47539)

Roberto L. Castro committed 4 months ago

[Bugfix][CPU] Ship examples/ in the CPU release image (#47447)

AgenticSpark committed 4 months ago

[Frontend] Refine the entrypoint class's inheritance hierarchy. (#47498)

wang.yuqi committed 4 months ago

[Bugfix][Frontend] Fix batch chat endpoint corrupting logprobs when return_token_ids is set (#47384)

David Feng committed 4 months ago