onnxruntime
[LARCH64 CPU]Provide inference acceleration optimization for Loongson CPU with 4-bit quantized models
#26280

Merged

[LARCH64 CPU]Provide inference acceleration optimization for Loongson CPU with 4-bit quantized models #26280

snnn merged 6 commits into microsoft:main from movedancer:add-more-loongson-support

This submission includes five modifications, which are as follows:

2ce037c4

snnn commented on 2025-10-14

movedancer requested a review from

snnn 255 days ago

Update sqnbitgemm_kernel_lasx_common.h

b3407de5

Update sqnbitgemm_kernel_lasx.cpp

7f841963

snnn commented on 2025-10-14

Update sqnbitgemm_kernel_lasx_common.h

4cae30ac

Update sqnbitgemm_kernel_lasx.cpp

a4c9b718

snnn closed this 253 days ago

snnn reopened this 253 days ago

Add MlasAlignedAllocator for aligned memory allocation

0a99f583

snnn approved these changes on 2025-10-17

snnn merged 7cc28b0b into main 251 days ago

movedancer deleted the add-more-loongson-support branch 251 days ago

Reviewers

snnn

Assignees

No one assigned

Labels

None yet

Milestone

No milestone