extensions
Module to load C++ / CUDA extensions.
Functions
Returns the cuda extention for tensor_quant. |
|
Returns the cuda extention for tensor_quant_fp8. |
|
Returns the cuda extention for tensor_quant_mx. |
|
Precompile the CUDA extensions. |
- get_cuda_ext()
Returns the cuda extention for tensor_quant.
- get_cuda_ext_fp8()
Returns the cuda extention for tensor_quant_fp8.
- get_cuda_ext_mx()
Returns the cuda extention for tensor_quant_mx.
- precompile()
Precompile the CUDA extensions.