diff --git a/src/plugins/intel_gpu/src/graph/impls/onednn/deconvolution_onednn.cpp b/src/plugins/intel_gpu/src/graph/impls/onednn/deconvolution_onednn.cpp
index 1d5707194c560d..ee2e80e57aca50 100644
--- a/src/plugins/intel_gpu/src/graph/impls/onednn/deconvolution_onednn.cpp
+++ b/src/plugins/intel_gpu/src/graph/impls/onednn/deconvolution_onednn.cpp
@@ -35,6 +35,14 @@ static std::shared_ptr<dnnl::deconvolution_forward::primitive_desc> get_deconvol
     auto output_md = onednn::layout_to_memory_desc(output_layout, tag_in_out);
     auto grouped_weights = format::is_grouped(weights_layout.format) || prim->grouped_weights_shape;
 
+    // Extend deconv parameters in case if spatials rank of output memory doesn't match size of parameters
+    int64_t insert_count = static_cast<int64_t>(output_md.get_dims().size()) - 2 - stride.size();
+    if (insert_count > 0) {
+        stride.insert(stride.end(), insert_count, 1);
+        pad_l.insert(pad_l.end(), insert_count, 0);
+        pad_r.insert(pad_r.end(), insert_count, 0);
+    }
+
     for (size_t i = 0; i < dilation.size(); i++) {
         dilation[i]--;
         int weights_offset = (grouped_weights ? 3 : 2) + static_cast<int>(i);
@@ -45,14 +53,6 @@ static std::shared_ptr<dnnl::deconvolution_forward::primitive_desc> get_deconvol
         pad_r[i] = (is - 1) * stride[i] - os + kernel_range - pad_l[i];
     }
 
-    // Extend deconv parameters in case if spatials rank of output memory doesn't match size of parameters
-    int64_t insert_count = static_cast<int64_t>(output_md.get_dims().size()) - 2 - stride.size();
-    if (insert_count > 0) {
-        stride.insert(stride.end(), insert_count, 1);
-        pad_l.insert(pad_l.end(), insert_count, 0);
-        pad_r.insert(pad_r.end(), insert_count, 0);
-    }
-
     if (!prim->bias.empty()) {
         auto bias_md = onednn::layout_to_memory_desc(impl_params.get_input_layout(2), dnnl::memory::format_tag::any, true);
         return std::make_shared<dnnl::deconvolution_forward::primitive_desc>(