From b9421898b65c48dc36259b1a434e855e6033fcf4 Mon Sep 17 00:00:00 2001 From: arthw Date: Thu, 23 Apr 2026 15:33:19 +0800 Subject: [PATCH] add for Q4_0 --- ggml/src/ggml-sycl/ggml-sycl.cpp | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/ggml/src/ggml-sycl/ggml-sycl.cpp b/ggml/src/ggml-sycl/ggml-sycl.cpp index e7f1c9ba5f..fc18014112 100644 --- a/ggml/src/ggml-sycl/ggml-sycl.cpp +++ b/ggml/src/ggml-sycl/ggml-sycl.cpp @@ -3624,10 +3624,13 @@ static void ggml_sycl_mul_mat(ggml_backend_sycl_context & ctx, const ggml_tensor if (!g_ggml_sycl_prioritize_dmmv && ((should_reorder_tensor(ctx, dst) && ggml_sycl_supports_reorder_mmvq(src0->type)))) { - //Arc770 get benefit with Q4_0 by skipping it. - if (ggml_sycl_info().devices[ctx.device].hw_info.arch != gpu_arch::intel_gpu_acm_g10) { - use_dequantize_mul_mat_vec = use_dequantize_mul_mat_vec && !use_mul_mat_vec_q; - } + // Arc770 get benefit with Q4_0 by skipping it. + if (!(ggml_sycl_info().devices[ctx.device].hw_info.arch == + gpu_arch::intel_gpu_acm_g10 && + src0->type == GGML_TYPE_Q4_0)) { + use_dequantize_mul_mat_vec = + use_dequantize_mul_mat_vec && !use_mul_mat_vec_q; + } } if (!split && src0->type == GGML_TYPE_F16 && ggml_is_permuted(src0) && ggml_is_permuted(src1) && src1->ne[1] == 1) {