AWQ: Up to 2.66x higher throughput (#2566)
This commit is contained in:
@@ -70,6 +70,14 @@ torch::Tensor awq_gemm(
|
||||
torch::Tensor _scaling_factors,
|
||||
torch::Tensor _zeros,
|
||||
int split_k_iters);
|
||||
|
||||
torch::Tensor awq_dequantize(
|
||||
torch::Tensor _kernel,
|
||||
torch::Tensor _scaling_factors,
|
||||
torch::Tensor _zeros,
|
||||
int split_k_iters,
|
||||
int thx,
|
||||
int thy);
|
||||
#endif
|
||||
|
||||
void squeezellm_gemm(
|
||||
|
||||
Reference in New Issue
Block a user