[CI] Build macOS arm64 CPU wheel natively on the macmini queue (#48289)

Michael Goin committed 3 months ago

[Reasoning] Optimize TPOT for thinking budget when used with speculative decoding (#46662)

rishitdholakia13 committed 3 months ago

[ROCm] Retune MI355 selective_state_update float32 config on the unified effective_batch grid (#48373)

vanshbhatia-amd committed 3 months ago

[KV Offload] Split cpu_cache_usage_perc into write/read usage gauges (#47666)

Srinivas Krovvidi committed 3 months ago

[Doc] Sync four function docstrings with their signatures (#45437)

Daoyuan Li committed 3 months ago

[Bugfix] Make MLA+SWA check the layer's backend, not the model config (#48520)

Michael Goin committed 3 months ago

[CI Bug] Fully solve accuracy issue for DSv3.2 + MTP + Sequence Parallel (#48036)

Wentao Ye committed 3 months ago

[Bugfix] Gemma4 parser: classify channel-less output consistently in streaming and non-streaming (#48262)

adhi29 committed 3 months ago

[Bugfix] Include inline per-token-head scales in offloaded page transfer width (#48411)

Itay Etelis committed 3 months ago

[NIXL] Avoid reading expired blocks in bidirectional turn-2 read (#47021)

tomerg-nvidia committed 3 months ago

[Model] Enable LoRA support for tower and connector in LlavaNextVideo (#48594)

karthik committed 3 months ago

[Bugfix][Security] Fix concurrent sparse invariant race bypassing CVE remediation (#48583)

Juan Pérez de Algaba committed 3 months ago

[Misc] Rename VLLM_TRITON_ATTN_USE_TD to VLLM_TRITON_USE_TD (#45781)

Artur Fierka committed 4 months ago

[Test] Enable KV cache events for HMA models in CPU offloading test (#47754)

Itay Etelis committed 4 months ago

[XPU][CI] Add tests/v1/e2e/general/test_correctness_sliding_window.py in Intel GPU CI (#47231)

xiangdong committed 4 months ago

[Bugfix] Return 400 instead of 500 when multimodal data is sent to a text-only model (#48473)

Hoang Nguyen committed 4 months ago

Add Cosmos3 Edge Reasoner model (#48291)

adsridhar committed 4 months ago

fix(security): guard lm-format-enforcer regex compile with timeout (#47595)

Juan Pérez de Algaba committed 4 months ago

[Perf][Feat] Add generic cuteDSL LL BF16 router (GEMM) (#42562)

Roberto L. Castro committed 4 months ago

[ROCm] Retune MI355 selective_state_update float16 config on the unified effective_batch grid (#48372)

vanshbhatia-amd committed 4 months ago