diff --git a/ggml/src/ggml-cuda/mmvq.cu b/ggml/src/ggml-cuda/mmvq.cu index 19a72afa..76126d76 100644 --- a/ggml/src/ggml-cuda/mmvq.cu +++ b/ggml/src/ggml-cuda/mmvq.cu @@ -526,6 +526,7 @@ static void ggml_cuda_op_mul_mat_vec_q_impl(ggml_backend_cuda_context & ctx, ggm break; case GGML_TYPE_IQ4_KSS: mul_mat_vec_iq4_kss_q8_1_cuda(src0_dd_i, src1_ddq_i, dst_dd_i, ids_data, ne00, row_diff, src1_padded_row_size, src1_ncols, nrows_dst, ne2, nb02, nb12, nb2, ids_nb0, stream); + break; case GGML_TYPE_IQ4_KT: mul_mat_vec_iq4_kt_q8_1_cuda(src0_dd_i, src1_ddq_i, dst_dd_i, ids_data, ne00, row_diff, src1_padded_row_size, src1_ncols, nrows_dst, ne2, nb02, nb12, nb2, ids_nb0, stream); break;