adasdadsd
/

smallq-flash-attention-ascend

flash-attention

speculative-decoding

Model card Files Files and versions

smallq-flash-attention-ascend / tests /test_aclnn.cpp

Commit History

Replace with cleaned vector-only implementation

86d4a3b
verified

adasdadsd commited on May 12

v2.0: Full float32 precision pipeline, headDim=64 support, 448/448 production tests passed

2145fce
verified

adasdadsd commited on May 9

Delete tests/test_aclnn.cpp with huggingface_hub

a8cb3cf
verified

adasdadsd commited on May 9

Upload folder using huggingface_hub

86e9bfe
verified

adasdadsd commited on May 9