[Bugfix][Spec Decode] Size DFlash query buffers for cudagraph-padded batches (#50065)

Siddhant Bharti committed 3 months ago

[Rust Frontend] Extract shared tracing setup logic into `vllm-tracing` (#50129)

Bugen Zhao committed 3 months ago

[BugFix] Fix clang spinloop mwaitx include (#45532)

johnnyychiu committed 3 months ago

[ROCm][Bugfix] Sanitize AITER paged-MQA logits before sparse top-k for DeepSeek-V4 (#49714)

Shanshan Shen committed 3 months ago

[Bugfix][Multimodal] Fix video temporal padding estimates (#49030)

labAxiaoming committed 3 months ago

[compressed-tensors] update `find_matched_target` order to prioritize fused name matches over class match (#49483)

Brian Dellabetta committed 3 months ago

[ROCm] Fix and optimize GPT-J-style MRoPE (#49906)

Andreas Karatzas committed 3 months ago

[ROCm] Cache fp32 upcast of static e8m0 weight scale in AITER scaled_mm (#47773)

jiacao-amd committed 3 months ago

[Rust][Benchmark] Make `vllm bench serve` Rust delegation opt-in (#50081)

Bugen Zhao committed 3 months ago

[Attention] Skip sparse indexer scoring for dense short prefills (#48407)

Yiliu Dong committed 3 months ago

[Bugfix] Don't reuse engine core payload buffer while zmq is sending it (#50053)

Nick Hill committed 3 months ago

[Docs] Expand llm-d integration page (#45432)

IBRAHIM IBRAHIM committed 3 months ago

[Bugfix][KV Offload][P2P] Scope serve state to fetch rounds (#49877)

Itay Etelis committed 3 months ago

[Core] Warm up runner-owned Triton kernels before the first request (#49903)

Nick Hill committed 3 months ago

[Bugfix][KV Offload] Keep Mamba block span unscaled under DCP (#49964)

Jonguk Cheong committed 3 months ago

[Bugfix][Kernel] Fix batch invariance in RMSNorm kernels by pinning block size (#48391)

oops-oom committed 3 months ago

[CI] Wire untethered test files into CI jobs (#49340)

Nick Hill committed 3 months ago

[Bugfix] Add missing `vllm/models/kimi_k3/__init__.py` (#50131)

Harry Mellor committed 3 months ago

[KV Connector] Support NIXL heterogeneous P/D block sizes for hybrid models (#49612)

Nick Hill committed 3 months ago

add epilogue hook to flex attention (#45841)

liangel-02 committed 3 months ago