gemm_awq.h 150 Bytes
Newer Older
Samuel Tesfai's avatar
Samuel Tesfai committed
1
2
3
4
5
#pragma once

#include "common.h"
#include "Tensor.h"

Muyang Li's avatar
Muyang Li committed
6
Tensor awq_gemm_forward_cuda(Tensor _in_feats, Tensor _kernel, Tensor _scales, Tensor _zeros);