[Bugfix] Voxtral prompt/audio placeholder alignment (#34140)

Artus Krohn-Grimberghe committed 9 months ago

[Bugfix] Avoid duplicate k-proj weight emission in helper (#34142)

Artus Krohn-Grimberghe committed 9 months ago

[Kernel] use flashinfer for gdn prefill (#32846)

Jiangyun Zhu committed 9 months ago

[Bugfix] Fix shared expert input for latent MoE in EP+DP (Nemotron-H) (#34087)

TomerBN-Nvidia committed 9 months ago

[Kernel] FlashInfer: switch allreduce fusion to unified API (#33985)

Mohammad Miadh Angkad committed 9 months ago

Add NUMA Core binding in nixl_connector for CPU xPyD (#32365)

ZhengHongming888 committed 9 months ago

[CI][torch.compile] Fix incorrect filtering for E2E fusion tests on B200 (#34031)

Luka Govedič committed 9 months ago

[UX] Add `--language-model-only` for hybrid models (#34120)

Roger Wang committed 9 months ago

[Misc] Fix up attention benchmarks (#33810)

Lucas Wilkinson committed 9 months ago

[MODEL] Adding Support for Qwen3.5 Models (#34110)

JJJYmmm committed 9 months ago

[XPU][6/N] add xpu scaled_mm kernel (#34117)

zofia committed 9 months ago

[Fix] [CPU Backend] : Prepack weights for w8a8 oneDNN matmul (#33901)

Nikhil Gupta committed 9 months ago

[ASR] Fix audio benchmark and add RTFx metric (#32300)

Ekagra Ranjan committed 9 months ago

[CI] Remove empty image_size_factors for fuyu, glm4_1v, glm_ocr (#34107)

Andreas Karatzas committed 9 months ago

[Model] GLM adaptation (#34124)

Jee Jee Li committed 9 months ago

fix(cpu): fix mla_decode compilation on x86 without AVX512 (#34052)

ihb2032 committed 9 months ago

[BugFix] Fix `fastsafetensors` TP all procs using all GPUs (#34070)

Nick Hill committed 9 months ago

[Frontend][last/5] Make pooling entrypoints request schema consensus. (#31127)

wang.yuqi committed 9 months ago

[Tiny] Rename encoder budget file to more specific name (#34103)

Reagan Lee committed 9 months ago

[bug-fix] supported_tasks is breaking backward compatibility at init_app_state (#34027)

kourosh hakhamaneshi committed 9 months ago