Enable DeepSeek V4 and GLM-5.1 on SM120 (#43477)

Gabriel Wu committed 4 months ago

[v1][kvconnector] DecodeBenchConnector: fill list/tuple (Mamba/KDA) KV caches (#45080)

Dao007forever committed 4 months ago

[Quantization][CI] add humming lm-eval test (#43752)

Jinzhen Lin committed 4 months ago

[Bugfix][Spec Decode] Fix EAGLE drafter multimodal encoder cache misses (#46315)

Nick Hill committed 4 months ago

[EPLB] Enable nixl eplb communicator for elastic ep (#45013)

Ilya Markov committed 4 months ago

[Pooling] Validate non-negative rerank top_n (#46119)

Taneem Ibrahim committed 4 months ago

[MyPy] Fix mypy for `vllm/lora` (#41722)

Martin Hickey committed 4 months ago

[Bugfix][CPU] Fix CPU model runner v2 (#46365)

Li, Jiang committed 4 months ago

[Test][KV Offloading] Add unit tests for OffloadingSpecFactory and SecondaryTierFactory (#46355)

AlexHuang committed 4 months ago

[XPU] add awq format for INCXPULinear (#43404)

Liangliang Ma committed 4 months ago

[Doc] Document pull request limit (#46376)

Simon Mo committed 4 months ago

[Attention] Re-enable cross-layer KV cache layout for MLA via stride-aware kernels (#45111)

Yifan Qiao committed 4 months ago

[docs] link security docs from AGENTS (#46373)

Simon Mo committed 4 months ago

[XPU] update nixl to v1.2.0 (#46327)

liuzhenwei committed 4 months ago

[CPU][RISC-V] Add RVV micro GEMM for WNA16 (#44324)

wcy committed 4 months ago

[XPU][CI]Skip v1/spec_decode/test_speculators_correctness.py in intel GPU nightly (#46356)

xiangdong committed 4 months ago

[Bugfix] Reject matryoshka embedding dimensions above hidden size (#46313)

Maxwill Lin committed 4 months ago

[ROCm][DSV4] Disable TileLang MHC dispatch on gfx942 (#45931)

Tuukka Sarvi committed 4 months ago

[Model] ColQwen3.5: fix retrieval correctness (bias + bidirectional) (#46108)

Athrael Soju committed 4 months ago

[Bugfix] Avoid racy accepted counts in async spec decode (#45100)

Weiwei Sun committed 4 months ago