yuantao 8de4315229 Add support for openpangu_pro_moe_v2, which characterized by its different kv head size and sink kv in attention. ...

Signed-off-by: yuantao <2422264527@qq.com>

2025-11-15 12:00:40 +08:00

..

fixtures

…

language

[Config] Clean up SchedulerConfig initialization (#28665)

2025-11-14 22:41:02 +08:00

multimodal

Avoid bytecode hook and simplify TorchCompileWrapperWithCustomDipatch (#25110)

2025-11-14 14:11:10 -08:00

quantization

[Quantization] fix attention quantization of gpt_oss model (#27334)

2025-11-11 12:06:00 -05:00

__init__.py

…

registry.py

Add support for openpangu_pro_moe_v2, which characterized by its different kv head size and sink kv in attention.

2025-11-15 12:00:40 +08:00

test_initialization.py

[Attention] Refactor CUDA attention backend selection logic (#24794)

2025-11-11 07:40:44 -05:00

test_oot_registration.py

…

test_registry.py

…

test_terratorch.py

…

test_transformers.py

[Docs] Update the name of Transformers backend -> Transformers modeling backend (#28725)

2025-11-14 16:34:14 +00:00

test_utils.py

…

test_vision.py

…

utils.py

…