Yong Hoon Shin 98c89e16ff

Make key optional for rotary embedding (#17566) ...

Signed-off-by: Yong Hoon Shin <yhshin@meta.com>

2025-05-07 00:11:46 -07:00

..

guided_decoding

[Feature][Frontend]: Deprecate --enable-reasoning (#17452)

2025-05-01 06:46:16 -07:00

layers

Make key optional for rotary embedding (#17566)

2025-05-07 00:11:46 -07:00

model_loader

Add NeuronxDistributedInference support, Speculative Decoding, Dynamic on-device sampling (#16357)

2025-05-07 00:07:30 -07:00

models

[Kernel] Use fused rmsnorm for some models like qwen3 series (#17735)

2025-05-06 23:10:02 -07:00

__init__.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

custom_op.py

[Fix] Support passing args to logger (#17425)

2025-04-30 08:06:58 -07:00

parameter.py

[Kernel] Support Microsoft Runtime Kernel Lib for our Low Precision Computation - BitBLAS (#6036)

2025-04-22 09:01:36 +01:00

pooling_metadata.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

sampling_metadata.py

[Bugfix] Fix extra comma (#15851)

2025-03-31 22:57:28 -07:00

utils.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00