extensions

Module to load C++ / CUDA extensions.

Functions

get_cuda_ext

Returns the cuda extention for tensor_quant.

get_cuda_ext_fp8

Returns the cuda extention for tensor_quant_fp8.

precompile

Precompile the CUDA extensions.

get_cuda_ext(raise_if_failed=False)

Returns the cuda extention for tensor_quant.

Parameters:

raise_if_failed (bool) –

get_cuda_ext_fp8(raise_if_failed=False)

Returns the cuda extention for tensor_quant_fp8.

Parameters:

raise_if_failed (bool) –

precompile()

Precompile the CUDA extensions.