[CI][Fix] Resolved the Ascend NPU test build image fail and add file dependencies (#50504)

Zeyu Yang committed about 2 months ago

[CI] Shard entrypoints API-server tests (#52344)

Kevin H. Luu committed about 2 months ago

[CI] Batch the swap_blocks verification instead of copying block by block (#54558)

stefankoncarevic committed about 2 months ago

[CI] Shard H100 MoE refactor integration tests (#52352)

Kevin H. Luu committed about 2 months ago

[KV Connector][Offloading] Look through UniformTypeKVCacheSpecs in the canonical portability gate (#51690)

Itay Etelis committed about 2 months ago

[Frontend] Add site-packages support for reasoning/tool parser plugins (#45241)

Yusuke Oda committed about 2 months ago

[Kimi-K3] Bump FlashKDA to fix unstable inverse (#54859)

Thien Tran committed about 2 months ago

[Frontend] Gate scale-out endpoints behind opt-in flag (#54579)

Francisco Javier Arceo committed about 2 months ago

[Bugfix] Implicitly close DeepSeek DSML parameters (#54838)

Flora Feng committed about 2 months ago

[CI] Use PR head label for Buildkite branch to avoid main collision (#54895)

Kevin H. Luu committed about 2 months ago

[XPU][UT] skip fp8_per_channel test on XPU (#54861)

Qiming Zhang committed about 2 months ago

[ROCm][AMD][Installation] Add mooncake build to rocm base image (#52650)

Giuseppe Grossi committed about 2 months ago

[XPU][CI] Move heavy jobs to nightly test in Intel GPU CI (#54863)

xiangdong committed about 2 months ago

[CI][ROCm] Calibrate AMD test timeouts from nightly runtimes (#54695)

Andreas Karatzas committed about 2 months ago

[Qwen3.8-Flash-Next] Fuse Qwen4Exp PLE kernels (#54517)

Thien Tran committed about 2 months ago

[CI] Shard basic model initialization tests (#54753)

Kevin H. Luu committed about 2 months ago

[Bugfix] Lazy-import FlashInfer PCIe IPC all-reduce in kernel_warmup (#54869)

Gabriel Wu committed about 2 months ago

[Qwen4] validate FP8 PLE weight scale after loading (#54722)

Harry Huang committed about 2 months ago

[Performance] Register Triton W4A16 GEMM as a custom op (#51453)

Giuseppe Grossi committed about 2 months ago

[Bugfix] Fix RoPE construction for deepseek-v4 sparse SWA layers (#54815)

Isotr0py committed about 2 months ago