diff --git a/server/exllamav2_kernels/exllamav2_kernels/cuda/q_gemm.cu b/server/exllamav2_kernels/exllamav2_kernels/cuda/q_gemm.cu index 73de5681..b4e4cf22 100644 --- a/server/exllamav2_kernels/exllamav2_kernels/cuda/q_gemm.cu +++ b/server/exllamav2_kernels/exllamav2_kernels/cuda/q_gemm.cu @@ -135,7 +135,7 @@ void gemm_half_q_half_cuda if (!temp_dq) temp_dq = b->temp_dq; b->reconstruct(temp_dq); - + //cublasSetMathMode(cublas_handle, CUBLAS_TENSOR_OP_MATH); const half alpha = __float2half(1.0f);