diff --git a/samples/cpp/benchmark_app/main.cpp b/samples/cpp/benchmark_app/main.cpp
index 7e32df4320f..55c10c8f867 100644
--- a/samples/cpp/benchmark_app/main.cpp
+++ b/samples/cpp/benchmark_app/main.cpp
@@ -193,6 +193,21 @@ int main(int argc, char* argv[]) {
             slog::info << "GPU extensions is loaded " << ext << slog::endl;
         }
 
+        if (FLAGS_hint.empty()) {
+            for (auto& device : devices) {
+                std::vector<std::string> supported_config_keys =
+                    core.get_metric(device, METRIC_KEY(SUPPORTED_CONFIG_KEYS));
+                if (std::find(supported_config_keys.begin(),
+                              supported_config_keys.end(),
+                              CONFIG_KEY(PERFORMANCE_HINT)) != supported_config_keys.end()) {
+                    slog::warn << "-hint default value is determined as" << CONFIG_VALUE(THROUGHPUT)
+                               << " automatically for " << device
+                               << " device. For more detailed information look at README." << slog::endl;
+                    FLAGS_hint = CONFIG_VALUE(THROUGHPUT);
+                }
+            }
+        }
+
         slog::info << "OpenVINO: " << ov::get_openvino_version() << slog::endl;
         slog::info << "Device info: " << slog::endl;
         slog::info << core.get_versions(device_name) << slog::endl;
@@ -562,19 +577,16 @@ int main(int argc, char* argv[]) {
         // ----------------- 8. Querying optimal runtime parameters
         // -----------------------------------------------------
         next_step();
-        // output of the actual settings that the device selected based on the hint
-        if (!ov_perf_hint.empty()) {
-            for (const auto& device : devices) {
-                std::vector<std::string> supported_config_keys =
-                    core.get_metric(device, METRIC_KEY(SUPPORTED_CONFIG_KEYS));
-                slog::info << "Device: " << device << slog::endl;
-                for (const auto& cfg : supported_config_keys) {
-                    try {
-                        slog::info << "  {" << cfg << " , " << compiledModel.get_config(cfg).as<std::string>();
-                    } catch (...) {
-                    };
+        // output of the actual settings that the device selected
+        for (const auto& device : devices) {
+            std::vector<std::string> supported_config_keys = core.get_metric(device, METRIC_KEY(SUPPORTED_CONFIG_KEYS));
+            slog::info << "Device: " << device << slog::endl;
+            for (const auto& cfg : supported_config_keys) {
+                try {
+                    slog::info << "  {" << cfg << " , " << compiledModel.get_config(cfg).as<std::string>();
                     slog::info << " }" << slog::endl;
-                }
+                } catch (...) {
+                };
             }
         }
 
diff --git a/src/inference/include/ie/ie_core.hpp b/src/inference/include/ie/ie_core.hpp
index d41e85b8e2d..87de21937d7 100644
--- a/src/inference/include/ie/ie_core.hpp
+++ b/src/inference/include/ie/ie_core.hpp
@@ -79,6 +79,7 @@ public:
      * @return CNNNetwork
      */
     CNNNetwork ReadNetwork(const std::string& modelPath, const std::string& binPath = {}) const;
+
     /**
      * @brief Reads models from IR and ONNX formats
      * @param model string with model in IR or ONNX format
diff --git a/tools/benchmark_tool/openvino/tools/benchmark/main.py b/tools/benchmark_tool/openvino/tools/benchmark/main.py
index 88100871012..7b33621a029 100644
--- a/tools/benchmark_tool/openvino/tools/benchmark/main.py
+++ b/tools/benchmark_tool/openvino/tools/benchmark/main.py
@@ -80,6 +80,14 @@ def run(args):
             cldnn_config = config[GPU_DEVICE_NAME]['CONFIG_FILE']
             benchmark.add_extension(path_to_cldnn_config=cldnn_config)
 
+        if not args.perf_hint:
+            for device in devices:
+                supported_config_keys = benchmark.core.get_metric(device, 'SUPPORTED_CONFIG_KEYS')
+                if 'PERFORMANCE_HINT' in supported_config_keys:
+                    logger.warning(f"-hint default value is determined as 'THROUGHPUT' automatically for {device} device" +
+                                    "For more detailed information look at README.")
+                    args.perf_hint = "throughput"
+
         version = benchmark.get_version_info()
 
         logger.info(version)
@@ -128,7 +136,7 @@ def run(args):
             perf_counts = True if config[device]['PERF_COUNT'] == 'YES' else perf_counts
 
             ## high-level performance hints
-            if is_flag_set_in_command_line('hint'):
+            if is_flag_set_in_command_line('hint') or args.perf_hint:
                 config[device]['PERFORMANCE_HINT'] = args.perf_hint.upper()
                 if is_flag_set_in_command_line('nireq'):
                     config[device]['PERFORMANCE_HINT_NUM_REQUESTS'] = str(args.number_infer_requests)
@@ -308,13 +316,16 @@ def run(args):
 
         # --------------------- 8. Querying optimal runtime parameters --------------------------------------------------
         next_step()
-        if is_flag_set_in_command_line('hint'):
-            ## actual device-deduced settings for the hint
-            for device in devices:
-                keys = benchmark.core.get_metric(device, 'SUPPORTED_CONFIG_KEYS')
-                logger.info(f'DEVICE: {device}')
-                for k in keys:
-                    logger.info(f'  {k}  , {compiled_model.get_config(k)}')
+        ## actual device-deduced settings
+        for device in devices:
+            keys = benchmark.core.get_metric(device, 'SUPPORTED_CONFIG_KEYS')
+            logger.info(f'DEVICE: {device}')
+            for k in keys:
+                try:
+                    logger.info(f'  {k}  , {benchmark.core.get_config(device, k)}')
+                except:
+                    pass
+
 
         # Update number of streams
         for device in device_number_streams.keys():