[XPU] Route block-quantized FP8 weights to the W8A8 kernel (#50787)

Chaojun Zhang committed 2 months ago

[Bugfix][XPU] Run GDN attention as eager break under breakable cudagraph (#51928)

Huanxing committed 2 months ago

[Model] Support R3 capture with DeepGEMM MegaMoE (#51831)

aoshen02 committed 2 months ago

[Attention] Move context_lens_tensor compute into GDN prefill path (#51913)

Xin Yang committed 2 months ago

[CI/Release][XPU] fix workdir path for triton shim job (#51923)

Kunshang Ji committed 2 months ago

[Bugfix][Quantization] Reuse online NVFP4 MoE kernel across reloads (#50074)

Matej Sirovatka committed 2 months ago

[Refactor][MRV2] Unify uniform decode token count helper (#51917)

Lucas Wilkinson committed 2 months ago

[Bugfix] Avoid repeated layerwise reload warning scans (#49505)

aoshen02 committed 2 months ago

[Frontend] Make `model` optional on all `/derender` request classes (#51463)

Vinay R Damodaran committed 2 months ago

[XPU][CI]Change to use global VLLM_DISABLE_COMPILE_CACHE=1 in Intel GPU CI (#51905)

xiangdong committed 2 months ago

[Bugfix][V1][Multimodal] Recover from P0/P1 processor cache drift (#46747) (#46747)

WillZZZy committed 2 months ago

[CI/Release][1/N][XPU] Publish XPU Triton shim index (#51759)

Kunshang Ji committed 2 months ago

[ROCm][Perf] Kimi-K3 Fused kernel for KDA decode (#50654)

kliuae committed 2 months ago

[Bugfix][Triton] Make fp8_min/fp8_max constexpr in _quantize_pad_fp8_kernel (#51872)

Max Hu committed 2 months ago

[CI/Build] Use file rendezvous for local distributed tests (#51652)

Stellar鱼 committed 2 months ago

[Hardware][AMD] Enable fused bf16→fp32 router GEMM on ROCm (#50268)

Matvei Pashkovskii committed 2 months ago

[Perf] Avoid more GPU<->CPU syncs on the model execution path (#51738)

Nick Hill committed 2 months ago

[CI] Stabilize tensor IPC multiprocessing tests (#50804)

Andreas Karatzas committed 2 months ago

[Bugfix][TieredOffloading] : Return HIT_PENDING when KV promotion is triggered (#51840)

Varun Sundar Rabindranath committed 2 months ago

[Docs][RL] Rewrite weight-transfer docs; standardize examples (#51729)

Aaron Hao committed 2 months ago