diff --git a/modules/nvidia_plugin/include/nvidia/nvidia_config.hpp b/modules/nvidia_plugin/include/nvidia/nvidia_config.hpp
deleted file mode 100644
index ee501dfd6..000000000
--- a/modules/nvidia_plugin/include/nvidia/nvidia_config.hpp
+++ /dev/null
@@ -1,65 +0,0 @@
-// Copyright (C) 2018-2021 Intel Corporation
-// SPDX-License-Identifier: Apache-2.0
-//
-
-/**
- * @brief A header that defines advanced related properties for DLIA plugins.
- * These properties should be used in SetConfig() and LoadNetwork() methods of plugins
- *
- * @file nvidia_config.hpp
- */
-
-#pragma once
-
-#include <string>
-
-#include "ie_plugin_config.hpp"
-
-namespace InferenceEngine {
-
-namespace CUDAMetrics {
-
-/**
- * @def NVIDIA_METRIC_VALUE(name)
- * @brief Shortcut for defining Template metric values
- */
-#define NVIDIA_METRIC_VALUE(name) InferenceEngine::CUDAMetrics::name
-#define DECLARE_NVIDIA_METRIC_VALUE(name) static constexpr auto name = #name
-
-// ! [public_header:metrics]
-/**
- * @brief Defines whether current Template device instance supports hardware blocks for fast convolution computations.
- */
-DECLARE_NVIDIA_METRIC_VALUE(HARDWARE_CONVOLUTION);
-// ! [public_header:metrics]
-
-}  // namespace CUDAMetrics
-
-namespace CUDAConfigParams {
-
-/**
- * @def NVIDIA_CONFIG_KEY(name)
- * @brief Shortcut for defining Template device configuration keys
- */
-#define NVIDIA_CONFIG_KEY(name) InferenceEngine::CUDAConfigParams::_CONFIG_KEY(NVIDIA_##name)
-#define NVIDIA_CONFIG_VALUE(name) InferenceEngine::CUDAConfigParams::NVIDIA_##name
-
-#define DECLARE_NVIDIA_CONFIG_KEY(name) DECLARE_CONFIG_KEY(NVIDIA_##name)
-#define DECLARE_NVIDIA_CONFIG_VALUE(name) DECLARE_CONFIG_VALUE(NVIDIA_##name)
-
-DECLARE_NVIDIA_CONFIG_VALUE(YES);
-DECLARE_NVIDIA_CONFIG_VALUE(NO);
-
-/**
- * @brief Defines the number of throutput streams used by NVIDIA GPU plugin.
- */
-DECLARE_NVIDIA_CONFIG_VALUE(THROUGHPUT_AUTO);
-DECLARE_NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS);
-
-/**
- * @brief Defines if optimization should be run for CUDA libraries ("NVIDIA_YES", "NVIDIA_NO" - default).
- */
-DECLARE_NVIDIA_CONFIG_KEY(OPERATION_BENCHMARK);
-
-}  // namespace CUDAConfigParams
-}  // namespace InferenceEngine
diff --git a/modules/nvidia_plugin/src/cancellation_token.hpp b/modules/nvidia_plugin/src/cancellation_token.hpp
index 89abad86f..911d0b6ae 100644
--- a/modules/nvidia_plugin/src/cancellation_token.hpp
+++ b/modules/nvidia_plugin/src/cancellation_token.hpp
@@ -4,8 +4,6 @@
 
 #pragma once
 
-#include <ie_extension.h>
-
 #include <atomic>
 #include <error.hpp>
 #include <functional>
diff --git a/modules/nvidia_plugin/src/cuda/blas.hpp b/modules/nvidia_plugin/src/cuda/blas.hpp
index c35826393..6bda4eb9d 100644
--- a/modules/nvidia_plugin/src/cuda/blas.hpp
+++ b/modules/nvidia_plugin/src/cuda/blas.hpp
@@ -5,7 +5,6 @@
 #pragma once
 
 #include <cublas_v2.h>
-#include <ie_extension.h>
 
 #include "runtime.hpp"
 
diff --git a/modules/nvidia_plugin/src/cuda_compiled_model.cpp b/modules/nvidia_plugin/src/cuda_compiled_model.cpp
index 9bbf59063..aa69f3251 100644
--- a/modules/nvidia_plugin/src/cuda_compiled_model.cpp
+++ b/modules/nvidia_plugin/src/cuda_compiled_model.cpp
@@ -19,7 +19,6 @@
 #include "memory_manager/cuda_immutable_memory_block_builder.hpp"
 #include "memory_manager/cuda_memory_manager.hpp"
 #include "memory_manager/model/cuda_memory_model_builder.hpp"
-#include "nvidia/nvidia_config.hpp"
 #include "nvidia/properties.hpp"
 #include "openvino/runtime/exec_model_info.hpp"
 #include "openvino/runtime/internal_properties.hpp"
diff --git a/modules/nvidia_plugin/src/cuda_compiled_model.hpp b/modules/nvidia_plugin/src/cuda_compiled_model.hpp
index 804704d5b..46cdd47a3 100644
--- a/modules/nvidia_plugin/src/cuda_compiled_model.hpp
+++ b/modules/nvidia_plugin/src/cuda_compiled_model.hpp
@@ -32,7 +32,7 @@ class CompiledModel : public ov::ICompiledModel {
                   const Configuration& cfg,
                   const std::shared_ptr<ov::threading::ITaskExecutor>& wait_executor,
                   const std::shared_ptr<const ov::IPlugin>& plugin,
-                  bool loaded_from_cache = false);
+                  bool loaded_from_cache);
 
     ~CompiledModel();
 
diff --git a/modules/nvidia_plugin/src/cuda_config.cpp b/modules/nvidia_plugin/src/cuda_config.cpp
index 83fc29d99..552ee2f88 100644
--- a/modules/nvidia_plugin/src/cuda_config.cpp
+++ b/modules/nvidia_plugin/src/cuda_config.cpp
@@ -3,10 +3,10 @@
 //
 
 #include "cuda_config.hpp"
+#include "openvino/runtime/internal_properties.hpp"
 
 #include <fmt/format.h>
 
-#include <cpp_interfaces/interface/ie_internal_plugin_config.hpp>
 #include <error.hpp>
 #include <regex>
 
@@ -154,19 +154,6 @@ Configuration::Configuration(const ov::AnyMap& config, const Configuration& defa
 
         if (ov::num_streams == key) {
             num_streams = value.as<ov::streams::Num>();
-        } if (NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS) == key) {
-            if (value != NVIDIA_CONFIG_VALUE(THROUGHPUT_AUTO)) {
-                try {
-                    num_streams = value.as<ov::streams::Num>();
-                } catch (...) {
-                    throw_ov_exception(
-                        fmt::format("NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS) = {} "
-                                    "is not a number !!",
-                                    value.as<std::string>()));
-                }
-            } else {
-                num_streams = ov::streams::AUTO;
-            }
         } else if (ov::device::id == key) {
             // Device id is updated already
             continue;
@@ -215,11 +202,6 @@ ov::Any Configuration::get(const std::string& name) const {
     } else if (name == ov::num_streams) {
         return (num_streams == 0) ?
             ov::streams::Num(get_optimal_number_of_streams()) : num_streams;
-    } else if (name == NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS)) {
-        auto value = (num_streams == 0) ?
-            ov::streams::Num(get_optimal_number_of_streams()) : num_streams;
-        return (value ==  ov::streams::AUTO) ? NVIDIA_CONFIG_VALUE(THROUGHPUT_AUTO)
-                                             : ov::util::to_string(value);
     } else if (name == ov::hint::num_requests) {
         return hint_num_requests;
     } else if (name == ov::hint::inference_precision) {
diff --git a/modules/nvidia_plugin/src/cuda_config.hpp b/modules/nvidia_plugin/src/cuda_config.hpp
index 94cc18cdd..fafa84436 100644
--- a/modules/nvidia_plugin/src/cuda_config.hpp
+++ b/modules/nvidia_plugin/src/cuda_config.hpp
@@ -8,7 +8,6 @@
 #include <memory>
 #include <string>
 
-#include "nvidia/nvidia_config.hpp"
 #include "openvino/runtime/properties.hpp"
 #include "openvino/runtime/threading/istreams_executor.hpp"
 
diff --git a/modules/nvidia_plugin/src/cuda_infer_request.cpp b/modules/nvidia_plugin/src/cuda_infer_request.cpp
index 598501758..d0c118c85 100644
--- a/modules/nvidia_plugin/src/cuda_infer_request.cpp
+++ b/modules/nvidia_plugin/src/cuda_infer_request.cpp
@@ -8,7 +8,6 @@
 #include <fmt/format.h>
 
 #include <algorithm>
-#include <description_buffer.hpp>
 #include <gsl/span_ext>
 #include <map>
 #include <memory>
diff --git a/modules/nvidia_plugin/src/cuda_operation_base.cpp b/modules/nvidia_plugin/src/cuda_operation_base.cpp
index 717c55d44..2eebaa296 100644
--- a/modules/nvidia_plugin/src/cuda_operation_base.cpp
+++ b/modules/nvidia_plugin/src/cuda_operation_base.cpp
@@ -5,6 +5,8 @@
 #include <utility>
 
 #include "cuda_operation_base.hpp"
+#include "openvino/core/node.hpp"
+
 namespace ov {
 namespace nvidia_gpu {
 
@@ -14,8 +16,8 @@ OperationBase::OperationBase(const CreationContext& /*context*/,
                              IndexCollection&& outputIds)
     : node_name_{node.get_friendly_name()},
       type_name_{node.get_type_info().name},
-      input_ids_{move(inputIds)},
-      output_ids_{move(outputIds)} {
+      input_ids_{inputIds},
+      output_ids_{outputIds} {
     if (node.get_input_size() > 0) {
         runtime_precision_ = node.get_input_element_type(0);
     } else if (node.get_output_size() > 0) {
diff --git a/modules/nvidia_plugin/src/cuda_operation_base.hpp b/modules/nvidia_plugin/src/cuda_operation_base.hpp
index d95273175..09118458f 100644
--- a/modules/nvidia_plugin/src/cuda_operation_base.hpp
+++ b/modules/nvidia_plugin/src/cuda_operation_base.hpp
@@ -4,8 +4,6 @@
 
 #pragma once
 
-#include <ie_layouts.h>
-
 #include <cuda/device_pointers.hpp>
 #include <cuda_creation_context.hpp>
 #include <cuda_inference_request_context.hpp>
diff --git a/modules/nvidia_plugin/src/cuda_operation_registry.hpp b/modules/nvidia_plugin/src/cuda_operation_registry.hpp
index 89e5518e3..8af656eae 100644
--- a/modules/nvidia_plugin/src/cuda_operation_registry.hpp
+++ b/modules/nvidia_plugin/src/cuda_operation_registry.hpp
@@ -10,6 +10,7 @@
 #include <unordered_map>
 
 #include "cuda_operation_base.hpp"
+#include "openvino/core/node.hpp"
 
 namespace ov {
 namespace nvidia_gpu {
diff --git a/modules/nvidia_plugin/src/cuda_plugin.cpp b/modules/nvidia_plugin/src/cuda_plugin.cpp
index cda4e2ec1..828ff0f91 100644
--- a/modules/nvidia_plugin/src/cuda_plugin.cpp
+++ b/modules/nvidia_plugin/src/cuda_plugin.cpp
@@ -3,18 +3,15 @@
 //
 #include <fmt/format.h>
 
-#include "ie_metric_helpers.hpp"
-
-#include "cpp_interfaces/interface/ie_internal_plugin_config.hpp"
 #include "cuda/props.hpp"
 #include "cuda_compiled_model.hpp"
 #include "cuda_infer_request.hpp"
 #include "cuda_itt.hpp"
 #include "cuda_operation_registry.hpp"
 #include "cuda_plugin.hpp"
-#include "nvidia/nvidia_config.hpp"
 #include "openvino/core/op_extension.hpp"
 #include "openvino/op/util/op_types.hpp"
+#include "openvino/runtime/internal_properties.hpp"
 #include "openvino/runtime/core.hpp"
 #include "openvino/runtime/properties.hpp"
 #include "openvino/runtime/threading/executor_manager.hpp"
@@ -74,7 +71,8 @@ std::shared_ptr<ov::ICompiledModel> Plugin::compile_model(const std::shared_ptr<
     auto compiled_model = std::make_shared<CompiledModel>(model->clone(),
                                                           full_config,
                                                           wait_executor,
-                                                          shared_from_this());
+                                                          shared_from_this(),
+                                                          false);
     return compiled_model;
 }
 
@@ -105,13 +103,22 @@ std::shared_ptr<ov::ICompiledModel> Plugin::import_model(std::istream& model_str
 
     auto model = get_core()->read_model(xml_string, weights);
 
-    auto full_config = get_full_config(properties);
+    // check ov::loaded_from_cache property and erase it due to not needed any more.
+    auto _properties = properties;
+    const auto& it = _properties.find(ov::loaded_from_cache.name());
+    bool loaded_from_cache = false;
+    if (it != _properties.end()) {
+        loaded_from_cache = it->second.as<bool>();
+        _properties.erase(it);
+    }
+
+    auto full_config = get_full_config(_properties);
     auto wait_executor = get_stream_executor(full_config);
     auto compiled_model= std::make_shared<CompiledModel>(model,
                                                          full_config,
                                                          wait_executor,
                                                          shared_from_this(),
-                                                         true);
+                                                         loaded_from_cache);
     return compiled_model;
 }
 
@@ -193,8 +200,6 @@ void Plugin::set_property(const ov::AnyMap& properties) {
 }
 
 ov::Any Plugin::get_property(const std::string& name, const ov::AnyMap& properties) const {
-    using namespace InferenceEngine::CUDAMetrics;
-
     auto full_config = get_full_config(properties);
 
     if (ov::supported_properties == name) {
@@ -233,7 +238,7 @@ ov::Any Plugin::get_property(const std::string& name, const ov::AnyMap& properti
             ov::device::capability::EXPORT_IMPORT,
             ov::device::capability::FP32,
             ov::device::capability::FP16}};
-     } else if (ov::range_for_streams == name) {
+    } else if (ov::range_for_streams == name) {
         return decltype(ov::range_for_streams)::value_type{1, Configuration::reasonable_limit_of_streams};
     } else if (ov::range_for_async_infer_requests == name) {
         return decltype(ov::range_for_async_infer_requests)::value_type{1, 1, 1};
diff --git a/modules/nvidia_plugin/src/cuda_thread_pool.cpp b/modules/nvidia_plugin/src/cuda_thread_pool.cpp
index d36eeb6f9..cc24887cd 100644
--- a/modules/nvidia_plugin/src/cuda_thread_pool.cpp
+++ b/modules/nvidia_plugin/src/cuda_thread_pool.cpp
@@ -6,8 +6,6 @@
 
 #include <fmt/format.h>
 
-#include <details/ie_exception.hpp>
-
 #include "cuda_latch.hpp"
 
 namespace ov {
diff --git a/modules/nvidia_plugin/src/kernels/details/error.cpp b/modules/nvidia_plugin/src/kernels/details/error.cpp
index a72860e0c..853b0e4b3 100644
--- a/modules/nvidia_plugin/src/kernels/details/error.cpp
+++ b/modules/nvidia_plugin/src/kernels/details/error.cpp
@@ -4,7 +4,6 @@
 
 #include <fmt/format.h>
 
-#include <details/ie_exception.hpp>
 #include <error.hpp>
 #include <exception>
 
diff --git a/modules/nvidia_plugin/src/memory_manager/cuda_device_mem_block.cpp b/modules/nvidia_plugin/src/memory_manager/cuda_device_mem_block.cpp
index 72a304be0..7c5ced6e1 100644
--- a/modules/nvidia_plugin/src/memory_manager/cuda_device_mem_block.cpp
+++ b/modules/nvidia_plugin/src/memory_manager/cuda_device_mem_block.cpp
@@ -7,7 +7,6 @@
 #include <cuda_runtime_api.h>
 
 #include <cuda/runtime.hpp>
-#include <details/ie_exception.hpp>
 #include <iostream>
 
 namespace ov {
diff --git a/modules/nvidia_plugin/src/ops/convolution_components/convolution_cudnn_components.cpp b/modules/nvidia_plugin/src/ops/convolution_components/convolution_cudnn_components.cpp
index faf030981..5909c05bd 100644
--- a/modules/nvidia_plugin/src/ops/convolution_components/convolution_cudnn_components.cpp
+++ b/modules/nvidia_plugin/src/ops/convolution_components/convolution_cudnn_components.cpp
@@ -9,7 +9,6 @@
 
 #include <cuda_config.hpp>
 #include <openvino/core/except.hpp>
-#include <nvidia/nvidia_config.hpp>
 #include <ops/converters.hpp>
 
 namespace ov::nvidia_gpu::Convolution::Details {
diff --git a/modules/nvidia_plugin/src/ops/convolution_cudnn.cpp b/modules/nvidia_plugin/src/ops/convolution_cudnn.cpp
index 66d25ab31..cee0b9131 100644
--- a/modules/nvidia_plugin/src/ops/convolution_cudnn.cpp
+++ b/modules/nvidia_plugin/src/ops/convolution_cudnn.cpp
@@ -6,7 +6,6 @@
 
 #include <cudnn.h>
 
-#include <details/ie_exception.hpp>
 #include <openvino/core/except.hpp>
 
 #include "cuda/constant_factory.hpp"
diff --git a/modules/nvidia_plugin/src/ops/convolution_cudnn_be.cpp b/modules/nvidia_plugin/src/ops/convolution_cudnn_be.cpp
index d4557a936..2d4743bd2 100644
--- a/modules/nvidia_plugin/src/ops/convolution_cudnn_be.cpp
+++ b/modules/nvidia_plugin/src/ops/convolution_cudnn_be.cpp
@@ -7,7 +7,6 @@
 #include <fmt/format.h>
 
 #include <algorithm>
-#include <details/ie_exception.hpp>
 #include <openvino/core/except.hpp>
 #include <ops/converters.hpp>
 
diff --git a/modules/nvidia_plugin/src/ops/details/cuda_ie_api_import_fix.hpp b/modules/nvidia_plugin/src/ops/details/cuda_ie_api_import_fix.hpp
deleted file mode 100644
index a58d02275..000000000
--- a/modules/nvidia_plugin/src/ops/details/cuda_ie_api_import_fix.hpp
+++ /dev/null
@@ -1,10 +0,0 @@
-// Copyright (C) 2021-2023 Intel Corporation
-// SPDX-License-Identifier: Apache-2.0
-//
-
-#pragma once
-
-#include <ie_api.h>
-
-#undef INFERENCE_ENGINE_DEPRECATED
-#define INFERENCE_ENGINE_DEPRECATED(msg) __attribute__((deprecated(msg)))
diff --git a/modules/nvidia_plugin/src/ops/details/cuda_ngraph_import_fix.hpp b/modules/nvidia_plugin/src/ops/details/cuda_ngraph_import_fix.hpp
deleted file mode 100644
index f7086f6bf..000000000
--- a/modules/nvidia_plugin/src/ops/details/cuda_ngraph_import_fix.hpp
+++ /dev/null
@@ -1,10 +0,0 @@
-// Copyright (C) 2021-2023 Intel Corporation
-// SPDX-License-Identifier: Apache-2.0
-//
-
-#pragma once
-
-#include <ngraph/deprecated.hpp>
-
-#undef NGRAPH_DEPRECATED
-#define NGRAPH_DEPRECATED(msg) __attribute__((deprecated(msg)))
diff --git a/modules/nvidia_plugin/src/ops/fused_convolution_cudnn.cpp b/modules/nvidia_plugin/src/ops/fused_convolution_cudnn.cpp
index 10f117d74..f103ece4e 100644
--- a/modules/nvidia_plugin/src/ops/fused_convolution_cudnn.cpp
+++ b/modules/nvidia_plugin/src/ops/fused_convolution_cudnn.cpp
@@ -6,7 +6,6 @@
 
 #include <cudnn.h>
 
-#include <details/ie_exception.hpp>
 #include <openvino/core/except.hpp>
 #include <ops/converters.hpp>
 
diff --git a/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_be.cpp b/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_be.cpp
index 9311177c9..c918434d9 100644
--- a/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_be.cpp
+++ b/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_be.cpp
@@ -8,7 +8,6 @@
 
 #include <algorithm>
 #include <cuda/constant_factory.hpp>
-#include <details/ie_exception.hpp>
 #include <openvino/core/except.hpp>
 #include <ops/converters.hpp>
 
diff --git a/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_decomposed.cpp b/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_decomposed.cpp
index 088a41790..9318a9d93 100644
--- a/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_decomposed.cpp
+++ b/modules/nvidia_plugin/src/ops/fused_convolution_cudnn_decomposed.cpp
@@ -6,7 +6,6 @@
 
 #include <cudnn.h>
 
-#include <details/ie_exception.hpp>
 #include <openvino/core/except.hpp>
 #include <ops/converters.hpp>
 
diff --git a/modules/nvidia_plugin/src/ops/logical_not.cpp b/modules/nvidia_plugin/src/ops/logical_not.cpp
index 97d121e75..8c1e7e44d 100644
--- a/modules/nvidia_plugin/src/ops/logical_not.cpp
+++ b/modules/nvidia_plugin/src/ops/logical_not.cpp
@@ -2,6 +2,7 @@
 // SPDX-License-Identifier: Apache-2.0
 //
 #include "logical_not.hpp"
+#include "openvino/core/except.hpp"
 
 #include <cuda_operation_registry.hpp>
 
diff --git a/modules/nvidia_plugin/src/transformer/cuda_graph_transformer.hpp b/modules/nvidia_plugin/src/transformer/cuda_graph_transformer.hpp
index 92c0f37d3..495d6f684 100644
--- a/modules/nvidia_plugin/src/transformer/cuda_graph_transformer.hpp
+++ b/modules/nvidia_plugin/src/transformer/cuda_graph_transformer.hpp
@@ -6,7 +6,6 @@
 
 #include <cuda/runtime.hpp>
 
-#include "cpp/ie_cnn_network.h"
 #include "cuda_config.hpp"
 #include "openvino/core/model.hpp"
 
diff --git a/modules/nvidia_plugin/tests/unit/compile_model.cpp b/modules/nvidia_plugin/tests/unit/compile_model.cpp
index 934145a17..7d76a602f 100644
--- a/modules/nvidia_plugin/tests/unit/compile_model.cpp
+++ b/modules/nvidia_plugin/tests/unit/compile_model.cpp
@@ -5,7 +5,6 @@
 #include <gtest/gtest.h>
 
 #include <memory>
-#include <nvidia/nvidia_config.hpp>
 #include <ops/matmul.hpp>
 #include <typeinfo>
 
@@ -95,10 +94,6 @@ INSTANTIATE_TEST_SUITE_P(CompileModelTest,
                          CompileModelTest::getTestCaseName);
 
 std::vector<PropertiesParams> num_streams_1_properties = {
-    {
-        {CONFIG_KEY(DEVICE_ID), "0"},
-        {NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS), "1"},
-    },
     {
         {ov::device::id.name(), "0"},
         {ov::num_streams.name(), "1"},
@@ -138,10 +133,6 @@ INSTANTIATE_TEST_SUITE_P(CompileModelTest,
                          CompileModelTest::getTestCaseName);
 
 std::vector<PropertiesParams> num_streams_8_properties = {
-    {
-        {CONFIG_KEY(DEVICE_ID), "0"},
-        {NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS), "8"},
-    },
     {
         {ov::device::id.name(), "0"},
         {ov::num_streams.name(), "8"},
@@ -177,11 +168,6 @@ INSTANTIATE_TEST_SUITE_P(CompileModelTest,
                          CompileModelTest::getTestCaseName);
 
 std::vector<PropertiesParams> num_streams_8_properties_exclusive = {
-    {
-        {CONFIG_KEY(DEVICE_ID), "0"},
-        {NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS), "8"},
-        {CONFIG_KEY(EXCLUSIVE_ASYNC_REQUESTS), CONFIG_VALUE(YES)},
-    },
     {
         {ov::device::id.name(), "0"},
         {ov::num_streams.name(), "8"},
@@ -220,10 +206,6 @@ INSTANTIATE_TEST_SUITE_P(CompileModelTest,
                          CompileModelTest::getTestCaseName);
 
 std::vector<PropertiesParams> num_streams_auto_properties = {
-    {
-        {CONFIG_KEY(DEVICE_ID), "0"},
-        {NVIDIA_CONFIG_KEY(THROUGHPUT_STREAMS), NVIDIA_CONFIG_VALUE(THROUGHPUT_AUTO)},
-    },
     {
         {ov::device::id.name(), "0"},
         {ov::num_streams.name(), ov::util::to_string(ov::streams::AUTO)},
diff --git a/modules/nvidia_plugin/tests/unit/memory_manager/model/cuda_memory_model_test.cpp b/modules/nvidia_plugin/tests/unit/memory_manager/model/cuda_memory_model_test.cpp
index b9410ef43..63f42a9ad 100644
--- a/modules/nvidia_plugin/tests/unit/memory_manager/model/cuda_memory_model_test.cpp
+++ b/modules/nvidia_plugin/tests/unit/memory_manager/model/cuda_memory_model_test.cpp
@@ -6,8 +6,6 @@
 
 #include <gtest/gtest.h>
 
-#include <details/ie_exception.hpp>
-
 TEST(MemoryModel, Empty) {
     using namespace ov::nvidia_gpu;