GRYS

Poor performance of zdot.

1

### Details The performance of the Zdot function is too low compared to other vector operations (axpy, vecmul). According to perf_math_kernel tests, several blas functions have the following performance results：...

Question: Error when substituting the quantized matrix multiplication operator.

2

In AWQ inference, the quantized weight matrix is dequantized to fp16 and then multiplied by the input matrix `x` in the linear layer. But I try to directly replace the...