diff --git a/src/plugins/intel_gpu/src/graph/graph_optimizer/select_preferred_formats.cpp b/src/plugins/intel_gpu/src/graph/graph_optimizer/select_preferred_formats.cpp
index 77b9154081a..9954c4416b7 100644
--- a/src/plugins/intel_gpu/src/graph/graph_optimizer/select_preferred_formats.cpp
+++ b/src/plugins/intel_gpu/src/graph/graph_optimizer/select_preferred_formats.cpp
@@ -39,6 +39,8 @@ void select_preferred_formats::run(program& p) {
         // Onednn primitive descriptor creation may fail, for example, due to asymmetric weight.
         try {
             if (n->is_type<convolution>()) {
+                if (n->as<convolution>().weights_zero_points_term())
+                    continue;
                 auto prim_desc = onednn::get_convolution_primitive_descriptor(*n->get_kernel_impl_params(),
                                                                               dnnl::primitive_attr(),
                                                                               dnnl::memory::format_tag::any);
diff --git a/src/plugins/intel_gpu/src/graph/layout_optimizer.cpp b/src/plugins/intel_gpu/src/graph/layout_optimizer.cpp
index f54fe8d46f0..36984b050ea 100644
--- a/src/plugins/intel_gpu/src/graph/layout_optimizer.cpp
+++ b/src/plugins/intel_gpu/src/graph/layout_optimizer.cpp
@@ -1083,7 +1083,7 @@ layout layout_optimizer::get_expected_layout(layout const& current_layout,
     bool use_onednn_impls = _optimization_attributes.use_onednn_impls && input_layout.data_type != data_types::f32;
     bool i8_u8_input = input_layout.data_type == data_types::u8 || input_layout.data_type == data_types::i8;
 
-    if (use_onednn_impls && onednn_valid_post_ops) {
+    if (use_onednn_impls && onednn_valid_post_ops && node.get_preferred_output_fmt() != format::any) {
         expected_format = node.get_preferred_output_fmt();
     } else {
         /* *************************** Native impls format selection part ************************** */