Skip to content

Connecting fp16xq4 gemm kernels (optimized for A100) to MatMulNBits<fp16> operator #1594

Connecting fp16xq4 gemm kernels (optimized for A100) to MatMulNBits<fp16> operator

Connecting fp16xq4 gemm kernels (optimized for A100) to MatMulNBits<fp16> operator #1594