liangsu9988's picture
Add package source and torch-universal build variant
b22e03e verified
|
Raw
History Blame Contribute Delete
317 Bytes
# csrc
CUDA extension placeholder for the FlexAttention replacement package.
The v1 package currently exposes the stable Python Tensor API and SDPA-backed
autograd fallback. Native CUDA forward/backward kernels should be added here
without changing the public functions in `torch-ext/flashrt_flex_attention_train`.