diff --git a/cgmanifests/generated/cgmanifest.json b/cgmanifests/generated/cgmanifest.json
index 5eeff2f5fc..87fe51851a 100644
--- a/cgmanifests/generated/cgmanifest.json
+++ b/cgmanifests/generated/cgmanifest.json
@@ -302,7 +302,7 @@
       "component": {
         "type": "git",
         "git": {
-          "commitHash": "369d6676423c2a6dbf4a5665c4b5010240d99d3c",
+          "commitHash": "ba6a4fb34fdeaa3613bf981610c657e7b663a699",
           "repositoryUrl": "https://github.com/onnx/onnx-tensorrt.git"
         },
         "comments": "onnx_tensorrt"
diff --git a/cmake/deps.txt b/cmake/deps.txt
index cf7e816a8b..d4310bcf4c 100644
--- a/cmake/deps.txt
+++ b/cmake/deps.txt
@@ -24,8 +24,8 @@ microsoft_wil;https://github.com/microsoft/wil/archive/5f4caba4e7a9017816e47becd
 mimalloc;https://github.com/microsoft/mimalloc/archive/refs/tags/v2.1.1.zip;d5ee7d34223d0567892db5179849939c8769dc41
 mp11;https://github.com/boostorg/mp11/archive/refs/tags/boost-1.79.0.zip;c8f04e378535ededbe5af52c8f969d2dedbe73d5
 onnx;https://github.com/onnx/onnx/archive/refs/tags/v1.13.1.zip;7a2517d3e7442109d540de2c86f29cd76d34cf28
-#use the commit where it's several commits after 8.5-GA branch (https://github.com/onnx/onnx-tensorrt/commit/369d6676423c2a6dbf4a5665c4b5010240d99d3c)
-onnx_tensorrt;https://github.com/onnx/onnx-tensorrt/archive/369d6676423c2a6dbf4a5665c4b5010240d99d3c.zip;62119892edfb78689061790140c439b111491275
+#use the last commit of 8.6-EA branch (https://github.com/onnx/onnx-tensorrt/commit/ba6a4fb34fdeaa3613bf981610c657e7b663a699)
+onnx_tensorrt;https://github.com/onnx/onnx-tensorrt/archive/ba6a4fb34fdeaa3613bf981610c657e7b663a699.zip;5a474ed86e2c4ee4085d3daeff8222866e933dc0
 protobuf;https://github.com/protocolbuffers/protobuf/archive/refs/tags/v21.12.zip;7cf2733949036c7d52fda017badcab093fe73bfa
 psimd;https://github.com/Maratyszcza/psimd/archive/072586a71b55b7f8c584153d223e95687148a900.zip;1f5454b01f06f9656b77e4a5e2e31d7422487013
 pthreadpool;https://github.com/Maratyszcza/pthreadpool/archive/1787867f6183f056420e532eec640cba25efafea.zip;e43e80781560c5ab404a4da20f34d846f5f5d101
diff --git a/onnxruntime/test/providers/cpu/math/topk_op_test.cc b/onnxruntime/test/providers/cpu/math/topk_op_test.cc
index 39d041e724..a41bdd61a4 100644
--- a/onnxruntime/test/providers/cpu/math/topk_op_test.cc
+++ b/onnxruntime/test/providers/cpu/math/topk_op_test.cc
@@ -591,7 +591,7 @@ static void top_3_explicit_aix_infinity(int opset_version, bool positive) {
   std::vector<int64_t> expected_indices = {0, 0, 1, 1, 2, 2};
   std::vector<int64_t> expected_dimensions = {3, 2};
   int64_t axis = 0;
-  RunTest(opset_version, 3, input_vals, input_dimensions, expected_vals, expected_indices, expected_dimensions, true, axis, 0, 1);
+  RunTest(opset_version, 3, input_vals, input_dimensions, expected_vals, expected_indices, expected_dimensions, false, axis, 0, 1);
 }
 
 TEST(TopKOperator, Top3ExplicitAxisInfinity) {
diff --git a/tools/ci_build/github/azure-pipelines/c-api-noopenmp-packaging-pipelines.yml b/tools/ci_build/github/azure-pipelines/c-api-noopenmp-packaging-pipelines.yml
index a1ef6fdd7c..8b4ebeddf6 100644
--- a/tools/ci_build/github/azure-pipelines/c-api-noopenmp-packaging-pipelines.yml
+++ b/tools/ci_build/github/azure-pipelines/c-api-noopenmp-packaging-pipelines.yml
@@ -123,7 +123,7 @@ stages:
     buildArch: x64
     msbuildPlatform: x64
     packageName: x64-tensorrt
-    buildparameter: --use_tensorrt --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --enable_onnx_tests --enable_wcos --build_java --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
+    buildparameter: --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --enable_onnx_tests --enable_wcos --build_java --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
     runTests: ${{ parameters.RunOnnxRuntimeTests }}
     buildJava: true
     java_artifact_id: onnxruntime_gpu
@@ -316,13 +316,13 @@ stages:
         Steps:
         - script: |
             tools/ci_build/get_docker_image.py \
-              --dockerfile tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_6_tensorrt8_5 \
+              --dockerfile tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6 \
               --context tools/ci_build/github/linux/docker \
               --docker-build-args "--network=host --build-arg POLICY=manylinux2014 --build-arg PLATFORM=x86_64 --build-arg DEVTOOLSET_ROOTPATH=/opt/rh/devtoolset-11/root --build-arg PREPEND_PATH=/opt/rh/devtoolset-11/root/usr/bin: --build-arg LD_LIBRARY_PATH_ARG=/opt/rh/devtoolset-11/root/usr/lib64:/opt/rh/devtoolset-11/root/usr/lib:/opt/rh/devtoolset-11/root/usr/lib64/dyninst:/opt/rh/devtoolset-11/root/usr/lib/dyninst:/usr/local/lib64 --build-arg BUILD_UID=$( id -u )" \
               --container-registry onnxruntimebuildcache \
               --multiple_repos \
-              --repository onnxruntimecuda116xtrt85build
-          displayName: "Get onnxruntimecuda116xtrt85build image for tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_6_tensorrt8_5"
+              --repository onnxruntimecuda118xtrt86build
+          displayName: "Get onnxruntimecuda118xtrt86build image for tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6"
           workingDirectory: $(Build.SourcesDirectory)/onnxruntime
         ContainerRegistry: onnxruntimebuildcache
 
@@ -374,7 +374,7 @@ stages:
       inputs:
         script: |
           docker run --gpus all -e CC=/opt/rh/devtoolset-11/root/usr/bin/cc -e CXX=/opt/rh/devtoolset-11/root/usr/bin/c++ -e CFLAGS="-Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fstack-protector-strong -fstack-clash-protection -fcf-protection -O3 -Wl,--strip-all" -e CXXFLAGS="-Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fstack-protector-strong -fstack-clash-protection -fcf-protection -O3 -Wl,--strip-all" -e NVIDIA_VISIBLE_DEVICES=all --rm --volume $(Build.SourcesDirectory):/src_dir \
-          --volume $(Build.ArtifactStagingDirectory):/artifact_src -e NIGHTLY_BUILD onnxruntimecuda116xtrt85build \
+          --volume $(Build.ArtifactStagingDirectory):/artifact_src -e NIGHTLY_BUILD onnxruntimecuda118xtrt86build \
           /src_dir/onnxruntime-inference-examples/c_cxx/squeezenet/run_capi_application.sh -o /src_dir/onnxruntime -p /artifact_src/onnxruntime-linux-x64-gpu-$(OnnxRuntimeVersion).tgz -w /src_dir/onnxruntime-inference-examples/c_cxx/squeezenet
         workingDirectory: '$(Build.ArtifactStagingDirectory)'
 
diff --git a/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-ci-pipeline.yml b/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-ci-pipeline.yml
index 969b520091..8391f009aa 100644
--- a/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-ci-pipeline.yml
+++ b/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-ci-pipeline.yml
@@ -22,10 +22,10 @@ jobs:
 
   - template: templates/get-docker-image-steps.yml
     parameters:
-      Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_6_tensorrt8_5
+      Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
       Context: tools/ci_build/github/linux/docker
       DockerBuildArgs: "--network=host --build-arg POLICY=manylinux2014 --build-arg PLATFORM=x86_64  --build-arg DEVTOOLSET_ROOTPATH=/opt/rh/devtoolset-11/root --build-arg PREPEND_PATH=/opt/rh/devtoolset-11/root/usr/bin: --build-arg LD_LIBRARY_PATH_ARG=/opt/rh/devtoolset-11/root/usr/lib64:/opt/rh/devtoolset-11/root/usr/lib:/opt/rh/devtoolset-11/root/usr/lib64/dyninst:/opt/rh/devtoolset-11/root/usr/lib/dyninst:/usr/local/lib64 --build-arg BUILD_UID=$( id -u )"
-      Repository: onnxruntimetensorrt85gpubuild
+      Repository: onnxruntimetensorrt86gpubuild
 
   - task: CmdLine@2
     inputs:
@@ -39,7 +39,7 @@ jobs:
             -e ALLOW_RELEASED_ONNX_OPSET_ONLY=0 \
             -e NIGHTLY_BUILD \
             -e BUILD_BUILDNUMBER \
-            onnxruntimetensorrt85gpubuild \
+            onnxruntimetensorrt86gpubuild \
             /opt/python/cp38-cp38/bin/python3 /onnxruntime_src/tools/ci_build/build.py \
               --build_dir /build --cmake_generator Ninja \
               --config Release \
@@ -47,9 +47,9 @@ jobs:
               --build_shared_lib \
               --parallel \
               --build_wheel \
-              --enable_onnx_tests --use_cuda --cuda_version=11.6 --cuda_home=/usr/local/cuda-11.6 --cudnn_home=/usr/local/cuda-11.6 \
+              --enable_onnx_tests --use_cuda --cuda_version=11.8 --cuda_home=/usr/local/cuda-11.8 --cudnn_home=/usr/local/cuda-11.8 \
               --enable_pybind --build_java \
-              --use_tensorrt --tensorrt_placeholder_builder --tensorrt_home /usr \
+              --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_placeholder_builder --tensorrt_home /usr \
               --cmake_extra_defines CMAKE_CUDA_HOST_COMPILER=/opt/rh/devtoolset-11/root/usr/bin/cc CMAKE_CUDA_ARCHITECTURES=75
       workingDirectory: $(Build.SourcesDirectory)
 
diff --git a/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-daily-perf-pipeline.yml b/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-daily-perf-pipeline.yml
index 2d93185718..ce34f902e6 100644
--- a/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-daily-perf-pipeline.yml
+++ b/tools/ci_build/github/azure-pipelines/linux-gpu-tensorrt-daily-perf-pipeline.yml
@@ -8,12 +8,13 @@ parameters:
 - name: TrtVersion
   displayName: TensorRT Version
   type: string
-  default: 8.5.1.1
+  default: 8.6.0.12
   values:
   - 8.2.1.8
   - 8.0.1.6
   - 8.4.1.5
   - 8.5.1.1
+  - 8.6.0.12
   - BIN
 
 - name: ModelGroups
diff --git a/tools/ci_build/github/azure-pipelines/templates/download-deps.yml b/tools/ci_build/github/azure-pipelines/templates/download-deps.yml
index 05607a4e8a..defa132ab4 100644
--- a/tools/ci_build/github/azure-pipelines/templates/download-deps.yml
+++ b/tools/ci_build/github/azure-pipelines/templates/download-deps.yml
@@ -11,7 +11,7 @@ steps:
       packageType: upack
       feed: '/7424c8e4-5c62-490e-95c4-79446f31017c'
       definition: '517c4f6f-5437-4392-a70d-4f15ec5be2f0'
-      version: 1.0.41
+      version: 1.0.44
       downloadPath: $(Build.BinariesDirectory)/deps
 
 # The private ADO project
@@ -22,7 +22,7 @@ steps:
       packageType: upack
       feed: '/4c7631f5-24c0-4307-8822-1aa8f180c325'
       definition: 'fd9dd5ad-b73e-4678-890e-edcf680dbc1a'
-      version: 1.0.41
+      version: 1.0.44
       downloadPath: $(Build.BinariesDirectory)/deps
 
 # You can add more ADO accounts at here.
diff --git a/tools/ci_build/github/azure-pipelines/templates/linux-gpu-tensorrt-packaging-pipeline.yml b/tools/ci_build/github/azure-pipelines/templates/linux-gpu-tensorrt-packaging-pipeline.yml
index 5f5dc00254..87e1d42d7a 100644
--- a/tools/ci_build/github/azure-pipelines/templates/linux-gpu-tensorrt-packaging-pipeline.yml
+++ b/tools/ci_build/github/azure-pipelines/templates/linux-gpu-tensorrt-packaging-pipeline.yml
@@ -26,17 +26,17 @@ stages:
     timeoutInMinutes:  180
     pool: 'Onnxruntime-Linux-GPU'
     variables:
-      CUDA_VERSION: '11.6'
+      CUDA_VERSION: '11.8'
     steps:
       - checkout: self
         clean: true
         submodules: recursive
       - template: get-docker-image-steps.yml
         parameters:
-          Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_6_tensorrt8_5
+          Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
           Context: tools/ci_build/github/linux/docker
           DockerBuildArgs: "--network=host --build-arg POLICY=manylinux2014 --build-arg PLATFORM=x86_64 --build-arg DEVTOOLSET_ROOTPATH=/opt/rh/devtoolset-11/root --build-arg PREPEND_PATH=/opt/rh/devtoolset-11/root/usr/bin: --build-arg LD_LIBRARY_PATH_ARG=/opt/rh/devtoolset-11/root/usr/lib64:/opt/rh/devtoolset-11/root/usr/lib:/opt/rh/devtoolset-11/root/usr/lib64/dyninst:/opt/rh/devtoolset-11/root/usr/lib/dyninst:/usr/local/lib64 --build-arg BUILD_UID=$( id -u )"
-          Repository: onnxruntimecuda114xtrt82build
+          Repository: onnxruntimecuda118xtrt86build
       - template: set-version-number-variables-step.yml
 
       - task: CmdLine@2
@@ -44,9 +44,9 @@ stages:
           script: |
             mkdir -p $HOME/.onnx
             docker run --gpus all -e CC=/opt/rh/devtoolset-11/root/usr/bin/cc -e CXX=/opt/rh/devtoolset-11/root/usr/bin/c++ -e CFLAGS="-Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fstack-protector-strong -fstack-clash-protection -fcf-protection -O3 -Wl,--strip-all" -e CXXFLAGS="-Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fstack-protector-strong -fstack-clash-protection -fcf-protection -O3 -Wl,--strip-all" -e NVIDIA_VISIBLE_DEVICES=all --rm --volume /data/onnx:/data/onnx:ro --volume $(Build.SourcesDirectory):/onnxruntime_src --volume $(Build.BinariesDirectory):/build \
-            --volume /data/models:/build/models:ro --volume $HOME/.onnx:/home/onnxruntimedev/.onnx -e NIGHTLY_BUILD onnxruntimecuda114xtrt82build \
+            --volume /data/models:/build/models:ro --volume $HOME/.onnx:/home/onnxruntimedev/.onnx -e NIGHTLY_BUILD onnxruntimecuda118xtrt86build \
             /opt/python/cp38-cp38/bin/python3 /onnxruntime_src/tools/ci_build/build.py --build_dir /build --config Release \
-            --skip_submodule_sync --parallel --build_shared_lib ${{ parameters.buildJavaOption }} --use_tensorrt --tensorrt_placeholder_builder --cuda_version=$(CUDA_VERSION) --cuda_home=/usr/local/cuda-$(CUDA_VERSION) --cudnn_home=/usr --tensorrt_home=/usr --cmake_extra_defines CMAKE_CUDA_HOST_COMPILER=/opt/rh/devtoolset-11/root/usr/bin/cc 'CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80'
+            --skip_submodule_sync --parallel --build_shared_lib ${{ parameters.buildJavaOption }} --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_placeholder_builder --cuda_version=$(CUDA_VERSION) --cuda_home=/usr/local/cuda-$(CUDA_VERSION) --cudnn_home=/usr --tensorrt_home=/usr --cmake_extra_defines CMAKE_CUDA_HOST_COMPILER=/opt/rh/devtoolset-11/root/usr/bin/cc 'CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80'
           workingDirectory: $(Build.SourcesDirectory)
 
       - ${{ if eq(parameters.buildJava, true) }}:
diff --git a/tools/ci_build/github/azure-pipelines/templates/py-linux-gpu.yml b/tools/ci_build/github/azure-pipelines/templates/py-linux-gpu.yml
index 7c1a8af644..d2a5d8f6dd 100644
--- a/tools/ci_build/github/azure-pipelines/templates/py-linux-gpu.yml
+++ b/tools/ci_build/github/azure-pipelines/templates/py-linux-gpu.yml
@@ -23,10 +23,10 @@ jobs:
 
     - template: get-docker-image-steps.yml
       parameters:
-        Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_6_tensorrt8_5
+        Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
         Context: tools/ci_build/github/linux/docker
         DockerBuildArgs: "--network=host --build-arg POLICY=manylinux2014 --build-arg PLATFORM=x86_64 --build-arg DEVTOOLSET_ROOTPATH=/opt/rh/devtoolset-11/root --build-arg PREPEND_PATH=/opt/rh/devtoolset-11/root/usr/bin: --build-arg LD_LIBRARY_PATH_ARG=/opt/rh/devtoolset-11/root/usr/lib64:/opt/rh/devtoolset-11/root/usr/lib:/opt/rh/devtoolset-11/root/usr/lib64/dyninst:/opt/rh/devtoolset-11/root/usr/lib/dyninst:/usr/local/lib64 --build-arg BUILD_UID=$( id -u ) --build-arg PLATFORM=${{ parameters.arch }}"
-        Repository: onnxruntimecuda116xtrt85build${{ parameters.arch }}
+        Repository: onnxruntimecuda118xtrt86build${{ parameters.arch }}
 
 
     - task: Bash@3
@@ -34,7 +34,7 @@ jobs:
       inputs:
         targetType: filePath
         filePath: tools/ci_build/github/linux/run_python_dockerbuild.sh
-        arguments: -i onnxruntimecuda116xtrt85build${{ parameters.arch }} -x "-d GPU"
+        arguments: -i onnxruntimecuda118xtrt86build${{ parameters.arch }} -x "-d GPU"
 
     - task: PublishBuildArtifacts@1
       displayName: 'Publish Artifact: ONNXRuntime python wheel'
diff --git a/tools/ci_build/github/azure-pipelines/templates/py-packaging-selectable-stage.yml b/tools/ci_build/github/azure-pipelines/templates/py-packaging-selectable-stage.yml
index 966ee17d4e..1522d1041f 100644
--- a/tools/ci_build/github/azure-pipelines/templates/py-packaging-selectable-stage.yml
+++ b/tools/ci_build/github/azure-pipelines/templates/py-packaging-selectable-stage.yml
@@ -307,10 +307,10 @@ stages:
 
       - template: get-docker-image-steps.yml
         parameters:
-          Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_6_tensorrt8_5
+          Dockerfile: tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
           Context: tools/ci_build/github/linux/docker
           DockerBuildArgs: "--network=host --build-arg POLICY=manylinux2014 --build-arg PLATFORM=x86_64 --build-arg DEVTOOLSET_ROOTPATH=/opt/rh/devtoolset-11/root --build-arg PREPEND_PATH=/opt/rh/devtoolset-11/root/usr/bin: --build-arg LD_LIBRARY_PATH_ARG=/opt/rh/devtoolset-11/root/usr/lib64:/opt/rh/devtoolset-11/root/usr/lib:/opt/rh/devtoolset-11/root/usr/lib64/dyninst:/opt/rh/devtoolset-11/root/usr/lib/dyninst:/usr/local/lib64 --build-arg BUILD_UID=$( id -u )"
-          Repository: onnxruntimecuda116xtrt85build
+          Repository: onnxruntimecuda118xtrt86build
 
       - task: CmdLine@2
         displayName: 'Build Python Wheel'
@@ -324,14 +324,14 @@ stages:
               --volume /data/models:/build/models:ro \
               --volume $HOME/.onnx:/home/onnxruntimedev/.onnx \
               -e BUILD_BUILDNUMBER \
-              onnxruntimecuda116xtrt85build \
+              onnxruntimecuda118xtrt86build \
                 $(PythonManylinuxDir)/bin/python3 /onnxruntime_src/tools/ci_build/build.py \
                   --build_dir /build --cmake_generator Ninja \
                   --config Release --update --build \
                   --skip_submodule_sync \
                   --parallel \
                   --build_wheel \
-                  --enable_onnx_tests --use_tensorrt --cuda_version=11.6 --tensorrt_home=/usr --cuda_home=/usr/local/cuda-11.6 --cudnn_home=/usr/local/cuda-11.6 \
+                  --enable_onnx_tests --use_tensorrt --use_tensorrt_builtin_parser --cuda_version=11.8 --tensorrt_home=/usr --cuda_home=/usr/local/cuda-11.8 --cudnn_home=/usr/local/cuda-11.8 \
                   ${{ parameters.gpu_build_py_parameters }} \
                   --cmake_extra_defines CMAKE_CUDA_HOST_COMPILER=/opt/rh/devtoolset-11/root/usr/bin/cc 'CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80'
           workingDirectory: $(Build.SourcesDirectory)
@@ -361,7 +361,7 @@ stages:
                   --skip_submodule_sync \
                   --parallel \
                   --build_wheel \
-                  --enable_onnx_tests --use_tensorrt --cuda_version=11.4 --tensorrt_home=/usr --cuda_home=/usr/local/cuda-11.4 --cudnn_home=/usr/local/cuda-11.4  \
+                  --enable_onnx_tests --use_tensorrt --use_tensorrt_builtin_parser --cuda_version=11.4 --tensorrt_home=/usr --cuda_home=/usr/local/cuda-11.4 --cudnn_home=/usr/local/cuda-11.4  \
                   ${{ parameters.gpu_build_py_parameters }} --ctest_path '' \
                   --cmake_extra_defines CMAKE_CUDA_HOST_COMPILER=/opt/rh/devtoolset-11/root/usr/bin/cc 'CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80'
 
@@ -392,7 +392,7 @@ stages:
       variables:
         CUDA_VERSION: '11.6'
         buildArch: x64
-        EpBuildFlags: --use_tensorrt --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=$(CUDA_VERSION) --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v$(CUDA_VERSION)" --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80"
+        EpBuildFlags: --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=$(CUDA_VERSION) --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v$(CUDA_VERSION)" --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80"
         EnvSetupScript: setup_env_gpu.bat
         EP_NAME: gpu
       strategy:
diff --git a/tools/ci_build/github/azure-pipelines/templates/py-packaging-stage.yml b/tools/ci_build/github/azure-pipelines/templates/py-packaging-stage.yml
index a094367567..89516d8c56 100644
--- a/tools/ci_build/github/azure-pipelines/templates/py-packaging-stage.yml
+++ b/tools/ci_build/github/azure-pipelines/templates/py-packaging-stage.yml
@@ -272,7 +272,7 @@ stages:
         parameters:
           MACHINE_POOL: 'onnxruntime-gpu-winbuild-t4'
           PYTHON_VERSION: '3.8'
-          EP_BUILD_FLAGS: --use_tensorrt --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
+          EP_BUILD_FLAGS: --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
           ENV_SETUP_SCRIPT: setup_env_gpu.bat
           EP_NAME: gpu
 
@@ -280,7 +280,7 @@ stages:
         parameters:
           MACHINE_POOL: 'onnxruntime-gpu-winbuild-t4'
           PYTHON_VERSION: '3.9'
-          EP_BUILD_FLAGS: --use_tensorrt --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
+          EP_BUILD_FLAGS: --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
           ENV_SETUP_SCRIPT: setup_env_gpu.bat
           EP_NAME: gpu
 
@@ -288,7 +288,7 @@ stages:
         parameters:
           MACHINE_POOL: 'onnxruntime-gpu-winbuild-t4'
           PYTHON_VERSION: '3.10'
-          EP_BUILD_FLAGS: --use_tensorrt --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
+          EP_BUILD_FLAGS: --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6"  --cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=52;60;61;70;75;80"
           ENV_SETUP_SCRIPT: setup_env_gpu.bat
           EP_NAME: gpu
 
diff --git a/tools/ci_build/github/azure-pipelines/win-gpu-tensorrt-ci-pipeline.yml b/tools/ci_build/github/azure-pipelines/win-gpu-tensorrt-ci-pipeline.yml
index ef55f4e4a1..ee2169cbc0 100644
--- a/tools/ci_build/github/azure-pipelines/win-gpu-tensorrt-ci-pipeline.yml
+++ b/tools/ci_build/github/azure-pipelines/win-gpu-tensorrt-ci-pipeline.yml
@@ -54,7 +54,7 @@ jobs:
     displayName: 'Generate cmake config'
     inputs:
       scriptPath: '$(Build.SourcesDirectory)\tools\ci_build\build.py'
-      arguments: '--config $(BuildConfig) --build_dir $(Build.BinariesDirectory) --skip_submodule_sync --build_shared_lib --update --cmake_generator "Visual Studio 16 2019" --build_wheel --enable_onnx_tests --use_tensorrt --tensorrt_placeholder_builder --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6" --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=75'
+      arguments: '--config $(BuildConfig) --build_dir $(Build.BinariesDirectory) --skip_submodule_sync --build_shared_lib --update --cmake_generator "Visual Studio 16 2019" --build_wheel --enable_onnx_tests --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_placeholder_builder --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6" --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=75'
       workingDirectory: '$(Build.BinariesDirectory)'
 
   - task: VSBuild@1
@@ -84,7 +84,7 @@ jobs:
      del wheel_filename_file
      python.exe -m pip install -q --upgrade %WHEEL_FILENAME%
      set PATH=$(Build.BinariesDirectory)\$(BuildConfig)\$(BuildConfig);%PATH%
-     python $(Build.SourcesDirectory)\tools\ci_build\build.py --config $(BuildConfig) --build_dir $(Build.BinariesDirectory) --skip_submodule_sync --build_shared_lib --test --cmake_generator "Visual Studio 16 2019" --build_wheel --enable_onnx_tests --use_tensorrt --tensorrt_placeholder_builder --tensorrt_home="C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6" --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=75
+     python $(Build.SourcesDirectory)\tools\ci_build\build.py --config $(BuildConfig) --build_dir $(Build.BinariesDirectory) --skip_submodule_sync --build_shared_lib --test --cmake_generator "Visual Studio 16 2019" --build_wheel --enable_onnx_tests --use_tensorrt --use_tensorrt_builtin_parser --tensorrt_placeholder_builder --tensorrt_home="C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8" --cuda_version=11.6 --cuda_home="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6" --cmake_extra_defines CMAKE_CUDA_ARCHITECTURES=75
 
     workingDirectory: '$(Build.BinariesDirectory)\$(BuildConfig)\$(BuildConfig)'
     displayName: 'Run tests'
diff --git a/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_4_tensorrt8_2 b/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_4_tensorrt8_2
deleted file mode 100644
index 4b2b889e53..0000000000
--- a/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_4_tensorrt8_2
+++ /dev/null
@@ -1,176 +0,0 @@
-ARG BASEIMAGE=nvidia/cuda:11.4.2-cudnn8-devel-centos7
-ARG POLICY=manylinux2014
-ARG PLATFORM=x86_64
-ARG DEVTOOLSET_ROOTPATH=
-ARG LD_LIBRARY_PATH_ARG=
-ARG PREPEND_PATH=
-
-#We need CUDA, TensorRT and manylinux. But the CUDA Toolkit End User License Agreement says NVIDIA CUDA Driver Libraries(libcuda.so, libnvidia-ptxjitcompiler.so) are only distributable in applications that meet this criteria:
-#1. The application was developed starting from a NVIDIA CUDA container obtained from Docker Hub or the NVIDIA GPU Cloud, and
-#2. The resulting application is packaged as a Docker container and distributed to users on Docker Hub or the NVIDIA GPU Cloud only.
-#So we use CUDA as the base image then add manylinux and TensorRT on top of it.
-
-#Build manylinux2014 docker image begin
-FROM $BASEIMAGE AS runtime_base
-ARG POLICY
-ARG PLATFORM
-ARG DEVTOOLSET_ROOTPATH
-ARG LD_LIBRARY_PATH_ARG
-ARG PREPEND_PATH
-LABEL maintainer="The ManyLinux project"
-
-ENV AUDITWHEEL_POLICY=${POLICY} AUDITWHEEL_ARCH=${PLATFORM} AUDITWHEEL_PLAT=${POLICY}_${PLATFORM}
-ENV LC_ALL=en_US.UTF-8 LANG=en_US.UTF-8 LANGUAGE=en_US.UTF-8
-ENV DEVTOOLSET_ROOTPATH=${DEVTOOLSET_ROOTPATH}
-ENV LD_LIBRARY_PATH=${LD_LIBRARY_PATH_ARG}
-ENV PATH=${PREPEND_PATH}${PATH}
-ENV PKG_CONFIG_PATH=/usr/local/lib/pkgconfig
-
-# first copy the fixup mirrors script, keep the script around
-COPY build_scripts/fixup-mirrors.sh /usr/local/sbin/fixup-mirrors
-
-# setup entrypoint, this will wrap commands with `linux32` with i686 images
-COPY build_scripts/install-entrypoint.sh \
-     build_scripts/update-system-packages.sh \
-     build_scripts/build_utils.sh \
-     /build_scripts/
-
-RUN /build_scripts/install-entrypoint.sh && rm -rf /build_scripts
-COPY manylinux-entrypoint /usr/local/bin/manylinux-entrypoint
-ENTRYPOINT ["manylinux-entrypoint"]
-
-COPY build_scripts/install-runtime-packages.sh \
-     build_scripts/update-system-packages.sh \
-     build_scripts/build_utils.sh \
-     /build_scripts/
-RUN manylinux-entrypoint /build_scripts/install-runtime-packages.sh && rm -rf /build_scripts/
-
-COPY build_scripts/build_utils.sh /build_scripts/
-
-COPY build_scripts/install-autoconf.sh /build_scripts/
-RUN export AUTOCONF_ROOT=autoconf-2.71 && \
-    export AUTOCONF_HASH=431075ad0bf529ef13cb41e9042c542381103e80015686222b8a9d4abef42a1c && \
-    export AUTOCONF_DOWNLOAD_URL=http://ftp.gnu.org/gnu/autoconf && \
-    manylinux-entrypoint /build_scripts/install-autoconf.sh
-
-COPY build_scripts/install-automake.sh /build_scripts/
-RUN export AUTOMAKE_ROOT=automake-1.16.5 && \
-    export AUTOMAKE_HASH=07bd24ad08a64bc17250ce09ec56e921d6343903943e99ccf63bbf0705e34605 && \
-    export AUTOMAKE_DOWNLOAD_URL=http://ftp.gnu.org/gnu/automake && \
-    manylinux-entrypoint /build_scripts/install-automake.sh
-
-COPY build_scripts/install-libtool.sh /build_scripts/
-RUN export LIBTOOL_ROOT=libtool-2.4.7 && \
-    export LIBTOOL_HASH=04e96c2404ea70c590c546eba4202a4e12722c640016c12b9b2f1ce3d481e9a8 && \
-    export LIBTOOL_DOWNLOAD_URL=http://ftp.gnu.org/gnu/libtool && \
-    manylinux-entrypoint /build_scripts/install-libtool.sh
-
-COPY build_scripts/install-libxcrypt.sh /build_scripts/
-RUN export LIBXCRYPT_VERSION=4.4.28 && \
-    export LIBXCRYPT_HASH=db7e37901969cb1d1e8020cb73a991ef81e48e31ea5b76a101862c806426b457 && \
-    export LIBXCRYPT_DOWNLOAD_URL=https://github.com/besser82/libxcrypt/archive && \
-    export PERL_ROOT=perl-5.34.0 && \
-    export PERL_HASH=551efc818b968b05216024fb0b727ef2ad4c100f8cb6b43fab615fa78ae5be9a && \
-    export PERL_DOWNLOAD_URL=https://www.cpan.org/src/5.0 && \
-    manylinux-entrypoint /build_scripts/install-libxcrypt.sh
-
-FROM runtime_base AS build_base
-COPY build_scripts/install-build-packages.sh /build_scripts/
-RUN manylinux-entrypoint /build_scripts/install-build-packages.sh
-
-
-FROM build_base AS build_git
-COPY build_scripts/build-git.sh /build_scripts/
-RUN export GIT_ROOT=git-2.36.2 && \
-    export GIT_HASH=6dc2cdea5fb23d823ba4871cc23222c1db31dfbb6d6c6ff74c4128700df57c68 && \
-    export GIT_DOWNLOAD_URL=https://www.kernel.org/pub/software/scm/git && \
-    manylinux-entrypoint /build_scripts/build-git.sh
-
-
-FROM build_base AS build_cpython
-COPY build_scripts/build-sqlite3.sh /build_scripts/
-RUN export SQLITE_AUTOCONF_ROOT=sqlite-autoconf-3390200 && \
-    export SQLITE_AUTOCONF_HASH=852be8a6183a17ba47cee0bbff7400b7aa5affd283bf3beefc34fcd088a239de && \
-    export SQLITE_AUTOCONF_DOWNLOAD_URL=https://www.sqlite.org/2022 && \
-    manylinux-entrypoint /build_scripts/build-sqlite3.sh
-
-COPY build_scripts/build-openssl.sh /build_scripts/
-RUN export OPENSSL_ROOT=openssl-1.1.1q && \
-    export OPENSSL_HASH=d7939ce614029cdff0b6c20f0e2e5703158a489a72b2507b8bd51bf8c8fd10ca && \
-    export OPENSSL_DOWNLOAD_URL=https://www.openssl.org/source && \
-    manylinux-entrypoint /build_scripts/build-openssl.sh
-
-COPY build_scripts/build-cpython.sh /build_scripts/
-
-
-FROM build_cpython AS build_cpython38
-COPY build_scripts/ambv-pubkey.txt /build_scripts/cpython-pubkeys.txt
-RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.8.13
-
-
-FROM build_cpython AS build_cpython39
-COPY build_scripts/ambv-pubkey.txt /build_scripts/cpython-pubkeys.txt
-RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.9.13
-
-
-FROM build_cpython AS build_cpython310
-COPY build_scripts/cpython-pubkey-310-311.txt /build_scripts/cpython-pubkeys.txt
-RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.10.5
-
-FROM build_cpython AS build_cpython311
-COPY build_scripts/cpython-pubkey-310-311.txt /build_scripts/cpython-pubkeys.txt
-RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.11.2
-
-FROM build_cpython AS all_python
-COPY build_scripts/install-pypy.sh \
-     build_scripts/pypy.sha256 \
-     build_scripts/finalize-python.sh \
-     /build_scripts/
-RUN manylinux-entrypoint /build_scripts/install-pypy.sh 3.8 7.3.9
-RUN manylinux-entrypoint /build_scripts/install-pypy.sh 3.9 7.3.9
-COPY --from=build_cpython38 /opt/_internal /opt/_internal/
-COPY --from=build_cpython39 /opt/_internal /opt/_internal/
-COPY --from=build_cpython310 /opt/_internal /opt/_internal/
-COPY --from=build_cpython311 /opt/_internal /opt/_internal/
-RUN manylinux-entrypoint /build_scripts/finalize-python.sh
-
-
-FROM runtime_base
-COPY --from=build_git /manylinux-rootfs /
-COPY --from=build_cpython /manylinux-rootfs /
-COPY --from=all_python /opt/_internal /opt/_internal/
-COPY build_scripts/finalize.sh \
-     build_scripts/update-system-packages.sh \
-     build_scripts/python-tag-abi-tag.py \
-     build_scripts/requirements3.8.txt \
-     build_scripts/requirements3.9.txt \
-     build_scripts/requirements3.10.txt \
-     build_scripts/requirements3.11.txt \
-     build_scripts/requirements-base-tools.txt \
-     /build_scripts/
-COPY build_scripts/requirements-tools/* /build_scripts/requirements-tools/
-RUN manylinux-entrypoint /build_scripts/finalize.sh && rm -rf /build_scripts
-
-ENV SSL_CERT_FILE=/opt/_internal/certs.pem
-
-CMD ["/bin/bash"]
-
-#Build manylinux2014 docker image end
-
-#Install TensorRT 8.2.2.1
-#RUN yum install -y wget
-RUN v="8.2.2-1.cuda11.4" &&\
-    yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo &&\
-    yum -y install libnvinfer8-${v} libnvparsers8-${v} libnvonnxparsers8-${v} libnvinfer-plugin8-${v} \
-        libnvinfer-devel-${v} libnvparsers-devel-${v} libnvonnxparsers-devel-${v} libnvinfer-plugin-devel-${v}
-
-#Add our own dependencies
-ADD scripts /tmp/scripts
-RUN cd /tmp/scripts && /tmp/scripts/manylinux/install_centos.sh && /tmp/scripts/manylinux/install_deps.sh && rm -rf /tmp/scripts
-
-ARG BUILD_UID=1001
-ARG BUILD_USER=onnxruntimedev
-RUN adduser --uid $BUILD_UID $BUILD_USER
-WORKDIR /home/$BUILD_USER
-USER $BUILD_USER
-ENV PATH /usr/local/dotnet:$PATH
diff --git a/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_4_tensorrt8_0 b/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
similarity index 91%
rename from tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_4_tensorrt8_0
rename to tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
index 12913678c6..6f12da4497 100644
--- a/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_4_tensorrt8_0
+++ b/tools/ci_build/github/linux/docker/Dockerfile.manylinux2014_cuda11_8_tensorrt8_6
@@ -1,4 +1,4 @@
-ARG BASEIMAGE=nvidia/cuda:11.4.0-cudnn8-devel-centos7
+ARG BASEIMAGE=nvidia/cuda:11.8.0-cudnn8-devel-centos7
 ARG POLICY=manylinux2014
 ARG PLATFORM=x86_64
 ARG DEVTOOLSET_ROOTPATH=
@@ -103,6 +103,11 @@ RUN export OPENSSL_ROOT=openssl-1.1.1q && \
 COPY build_scripts/build-cpython.sh /build_scripts/
 
 
+FROM build_cpython AS build_cpython37
+COPY build_scripts/cpython-pubkeys.txt /build_scripts/cpython-pubkeys.txt
+RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.7.13
+
+
 FROM build_cpython AS build_cpython38
 COPY build_scripts/ambv-pubkey.txt /build_scripts/cpython-pubkeys.txt
 RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.8.13
@@ -119,15 +124,17 @@ RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.10.5
 
 FROM build_cpython AS build_cpython311
 COPY build_scripts/cpython-pubkey-310-311.txt /build_scripts/cpython-pubkeys.txt
-RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.11.2
+RUN manylinux-entrypoint /build_scripts/build-cpython.sh 3.11.0b5
 
 FROM build_cpython AS all_python
 COPY build_scripts/install-pypy.sh \
      build_scripts/pypy.sha256 \
      build_scripts/finalize-python.sh \
      /build_scripts/
+RUN manylinux-entrypoint /build_scripts/install-pypy.sh 3.7 7.3.9
 RUN manylinux-entrypoint /build_scripts/install-pypy.sh 3.8 7.3.9
 RUN manylinux-entrypoint /build_scripts/install-pypy.sh 3.9 7.3.9
+COPY --from=build_cpython37 /opt/_internal /opt/_internal/
 COPY --from=build_cpython38 /opt/_internal /opt/_internal/
 COPY --from=build_cpython39 /opt/_internal /opt/_internal/
 COPY --from=build_cpython310 /opt/_internal /opt/_internal/
@@ -142,6 +149,7 @@ COPY --from=all_python /opt/_internal /opt/_internal/
 COPY build_scripts/finalize.sh \
      build_scripts/update-system-packages.sh \
      build_scripts/python-tag-abi-tag.py \
+     build_scripts/requirements3.7.txt \
      build_scripts/requirements3.8.txt \
      build_scripts/requirements3.9.txt \
      build_scripts/requirements3.10.txt \
@@ -157,12 +165,12 @@ CMD ["/bin/bash"]
 
 #Build manylinux2014 docker image end
 
-#Install TensorRT 8.0.1.6
+#Install TensorRT 8.6.0.12
 #RUN yum install -y wget
-RUN v="8.0.1-1.cuda11.3" &&\
+RUN v="8.6.0.12-1.cuda11.8" &&\
     yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo &&\
-    yum -y install libnvinfer8-${v} libnvparsers8-${v} libnvonnxparsers8-${v} libnvinfer-plugin8-${v} \
-        libnvinfer-devel-${v} libnvparsers-devel-${v} libnvonnxparsers-devel-${v} libnvinfer-plugin-devel-${v}
+    yum -y install libnvinfer8-${v} libnvparsers8-${v} libnvonnxparsers8-${v} libnvinfer-plugin8-${v} libnvinfer-vc-plugin8-${v}\
+        libnvinfer-devel-${v} libnvparsers-devel-${v} libnvonnxparsers-devel-${v} libnvinfer-plugin-devel-${v} libnvinfer-vc-plugin-devel-${v}
 
 #Add our own dependencies
 ADD scripts /tmp/scripts
@@ -173,4 +181,4 @@ ARG BUILD_USER=onnxruntimedev
 RUN adduser --uid $BUILD_UID $BUILD_USER
 WORKDIR /home/$BUILD_USER
 USER $BUILD_USER
-ENV PATH /usr/local/dotnet:$PATH
+ENV PATH /usr/local/dotnet:$PATH
\ No newline at end of file
diff --git a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_4_tensorrt8_0 b/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_4_tensorrt8_0
deleted file mode 100644
index c2c8259d48..0000000000
--- a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_4_tensorrt8_0
+++ /dev/null
@@ -1,52 +0,0 @@
-# --------------------------------------------------------------
-# Copyright (c) Microsoft Corporation. All rights reserved.
-# Licensed under the MIT License.
-# --------------------------------------------------------------
-# Dockerfile to run ONNXRuntime with TensorRT integration
-
-ARG OS_VERSION=18.04
-FROM nvidia/cuda:11.4.0-cudnn8-devel-ubuntu${OS_VERSION}
-
-# ONNX Runtime Variables
-ARG ONNXRUNTIME_REPO=https://github.com/Microsoft/onnxruntime
-ARG ONNXRUNTIME_BRANCH=main
-ARG CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80
-
-ENV PATH /usr/local/nvidia/bin:/usr/local/cuda/bin:/code/cmake-3.24.3-linux-x86_64/bin:/opt/miniconda/bin:${PATH}
-ENV LD_LIBRARY_PATH /opt/miniconda/lib:$LD_LIBRARY_PATH
-
-RUN apt-get update &&\
-    apt-get install -y sudo git bash unattended-upgrades wget
-RUN unattended-upgrade
-
-# Install python3
-RUN apt-get install -y --no-install-recommends \
-    python3 \
-    python3-pip \
-    python3-dev \
-    python3-wheel &&\
-    cd /usr/local/bin &&\
-    ln -s /usr/bin/python3 python &&\
-    ln -s /usr/bin/pip3 pip;
-
-RUN pip install --upgrade pip 
-RUN pip install setuptools>=41.0.0
-
-# Install TensorRT
-RUN v="8.0.1-1+cuda11.3" &&\
-    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub &&\
-    apt-get update &&\
-    sudo apt-get install -y libnvinfer8=${v} libnvonnxparsers8=${v} libnvparsers8=${v} libnvinfer-plugin8=${v} \
-        libnvinfer-dev=${v} libnvonnxparsers-dev=${v} libnvparsers-dev=${v} libnvinfer-plugin-dev=${v} \
-        python3-libnvinfer=${v}
-
-WORKDIR /code
-
-# Prepare onnxruntime repository & build onnxruntime with TensorRT
-RUN git clone --single-branch --branch ${ONNXRUNTIME_BRANCH} --recursive ${ONNXRUNTIME_REPO} onnxruntime &&\
-    /bin/sh onnxruntime/dockerfiles/scripts/install_common_deps.sh &&\
-    /bin/sh onnxruntime/dockerfiles/scripts/checkout_submodules.sh 8.0 &&\
-    cd onnxruntime &&\
-    /bin/sh build.sh --parallel --build_shared_lib --cuda_home /usr/local/cuda --cudnn_home /usr/lib/x86_64-linux-gnu/ --use_tensorrt --tensorrt_home /usr/lib/x86_64-linux-gnu/ --config Release --build_wheel --skip_tests --skip_submodule_sync --cmake_extra_defines '"CMAKE_CUDA_ARCHITECTURES='${CMAKE_CUDA_ARCHITECTURES}'"' &&\
-    pip install /code/onnxruntime/build/Linux/Release/dist/*.whl &&\
-    cd .. 
diff --git a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_4_tensorrt8_2 b/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_4_tensorrt8_2
deleted file mode 100644
index 61c4d4b6d7..0000000000
--- a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_4_tensorrt8_2
+++ /dev/null
@@ -1,55 +0,0 @@
-# --------------------------------------------------------------
-# Copyright (c) Microsoft Corporation. All rights reserved.
-# Licensed under the MIT License.
-# --------------------------------------------------------------
-# Dockerfile to run ONNXRuntime with TensorRT integration
-
-ARG OS_VERSION=20.04
-FROM nvidia/cuda:11.4.2-cudnn8-devel-ubuntu${OS_VERSION}
-
-
-# ONNX Runtime Variables
-ARG ONNXRUNTIME_REPO=https://github.com/Microsoft/onnxruntime
-ARG ONNXRUNTIME_BRANCH=main
-ARG CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80
-
-ENV PATH /usr/local/nvidia/bin:/usr/local/cuda/bin:/code/cmake-3.24.3-linux-x86_64/bin:/opt/miniconda/bin:${PATH}
-ENV LD_LIBRARY_PATH /opt/miniconda/lib:$LD_LIBRARY_PATH
-
-ENV DEBIAN_FRONTEND=noninteractive
-
-RUN apt-get update &&\
-    apt-get install -y sudo git bash unattended-upgrades wget
-RUN unattended-upgrade
-
-# Install python3
-RUN apt-get install -y --no-install-recommends \
-    python3 \
-    python3-pip \
-    python3-dev \
-    python3-wheel &&\
-    cd /usr/local/bin &&\
-    ln -s /usr/bin/python3 python &&\
-    ln -s /usr/bin/pip3 pip;
-
-RUN pip install --upgrade pip 
-RUN pip install setuptools>=41.0.0
-
-# Install TensorRT
-RUN v="8.2.1-1+cuda11.4" &&\
-    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub &&\
-    apt-get update &&\
-    sudo apt-get install -y libnvinfer8=${v} libnvonnxparsers8=${v} libnvparsers8=${v} libnvinfer-plugin8=${v} \
-        libnvinfer-dev=${v} libnvonnxparsers-dev=${v} libnvparsers-dev=${v} libnvinfer-plugin-dev=${v} \
-        python3-libnvinfer=${v}
-
-WORKDIR /code
-
-# Prepare onnxruntime repository & build onnxruntime with TensorRT
-RUN git clone --single-branch --branch ${ONNXRUNTIME_BRANCH} --recursive ${ONNXRUNTIME_REPO} onnxruntime &&\
-    /bin/sh onnxruntime/dockerfiles/scripts/install_common_deps.sh &&\
-    /bin/sh onnxruntime/dockerfiles/scripts/checkout_submodules.sh 8.2 &&\
-    cd onnxruntime &&\
-    /bin/sh build.sh --parallel --build_shared_lib --cuda_home /usr/local/cuda --cudnn_home /usr/lib/x86_64-linux-gnu/ --use_tensorrt --tensorrt_home /usr/lib/x86_64-linux-gnu/ --config Release --build_wheel --skip_tests --skip_submodule_sync --cmake_extra_defines '"CMAKE_CUDA_ARCHITECTURES='${CMAKE_CUDA_ARCHITECTURES}'"' &&\
-    pip install /code/onnxruntime/build/Linux/Release/dist/*.whl &&\
-    cd .. 
diff --git a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_8_tensorrt8_6 b/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_8_tensorrt8_6
new file mode 100644
index 0000000000..ecc532d3cb
--- /dev/null
+++ b/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_cuda11_8_tensorrt8_6
@@ -0,0 +1,82 @@
+# --------------------------------------------------------------
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+# --------------------------------------------------------------
+# Dockerfile to run ONNXRuntime with TensorRT integration
+
+# Build base image with required system packages
+FROM nvidia/cuda:11.8.0-cudnn8-devel-ubuntu20.04 AS base
+
+# The local directory into which to build and install CMAKE
+ARG ONNXRUNTIME_LOCAL_CODE_DIR=/code
+
+ENV PATH /usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/src/tensorrt/bin:${ONNXRUNTIME_LOCAL_CODE_DIR}/cmake-3.24.3-linux-x86_64/bin:/opt/miniconda/bin:${PATH}
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN apt-get update &&\
+    apt-get install -y sudo git bash unattended-upgrades wget
+RUN unattended-upgrade
+
+# Install python3
+RUN apt-get install -y --no-install-recommends \
+    python3 \
+    python3-pip \
+    python3-dev \
+    python3-wheel &&\
+    cd /usr/local/bin &&\
+    ln -s /usr/bin/python3 python &&\
+    ln -s /usr/bin/pip3 pip;
+
+RUN pip install --upgrade pip 
+RUN pip install setuptools>=41.0.0
+
+# Install TensorRT
+RUN v="8.6.0.12-1+cuda11.8" &&\
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub &&\
+    apt-get update &&\
+    sudo apt-get install -y libnvinfer8=${v} libnvonnxparsers8=${v} libnvparsers8=${v} libnvinfer-plugin8=${v} libnvinfer-vc-plugin8=${v}\
+        libnvinfer-dev=${v} libnvonnxparsers-dev=${v} libnvparsers-dev=${v} libnvinfer-plugin-dev=${v} libnvinfer-vc-plugin-dev=${v}\
+        python3-libnvinfer=${v} libnvinfer-samples=${v}
+
+# Compile trtexec
+RUN cd /usr/src/tensorrt/samples/trtexec && make
+
+
+# Build final image from base. Builds ORT.
+FROM base as final
+
+# ONNX Runtime arguments
+
+# URL to the github repo from which to clone ORT.
+ARG ONNXRUNTIME_REPO=https://github.com/Microsoft/onnxruntime
+
+# The local directory into which to clone ORT.
+ARG ONNXRUNTIME_LOCAL_CODE_DIR=/code
+
+# The git branch of ORT to checkout and build.
+ARG ONNXRUNTIME_BRANCH=main
+
+# Optional. The specific commit to pull and build from. If not set, the latest commit is used.
+ARG ONNXRUNTIME_COMMIT_ID
+
+# The supported CUDA architecture
+ARG CMAKE_CUDA_ARCHITECTURES=37;50;52;60;61;70;75;80
+
+WORKDIR ${ONNXRUNTIME_LOCAL_CODE_DIR}
+
+# Clone ORT repository with branch
+RUN git clone --single-branch --branch ${ONNXRUNTIME_BRANCH} --recursive ${ONNXRUNTIME_REPO} onnxruntime &&\
+    /bin/sh onnxruntime/dockerfiles/scripts/install_common_deps.sh
+
+WORKDIR ${ONNXRUNTIME_LOCAL_CODE_DIR}/onnxruntime
+
+# Reset to a specific commit if specified by build args.
+RUN if [ -z "$ONNXRUNTIME_COMMIT_ID" ] ; then echo "Building branch ${ONNXRUNTIME_BRANCH}" ;\
+    else echo "Building branch ${ONNXRUNTIME_BRANCH} @ commit ${ONNXRUNTIME_COMMIT_ID}" &&\
+    git reset --hard ${ONNXRUNTIME_COMMIT_ID} && git submodule update --recursive ; fi
+
+# Build ORT
+RUN /bin/sh build.sh --parallel --build_shared_lib --cuda_home /usr/local/cuda --cudnn_home /usr/lib/x86_64-linux-gnu/ --use_tensorrt --tensorrt_home /usr/lib/x86_64-linux-gnu/ --config Release --build_wheel --skip_tests --skip_submodule_sync --cmake_extra_defines '"CMAKE_CUDA_ARCHITECTURES='${CMAKE_CUDA_ARCHITECTURES}'"'
+
+# Intall ORT wheel
+RUN pip install ${ONNXRUNTIME_LOCAL_CODE_DIR}/onnxruntime/build/Linux/Release/dist/*.whl
diff --git a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_tensorrt b/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_tensorrt
index 900267b9bb..c3b41afecb 100644
--- a/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_tensorrt
+++ b/tools/ci_build/github/linux/docker/Dockerfile.ubuntu_tensorrt
@@ -1,6 +1,6 @@
-# Tag: nvidia/cuda:11.6.1-cudnn8-devel-ubuntu20.04
-# Label: com.nvidia.cuda.version: 11.6.1
-# Label: com.nvidia.cudnn.version: 8.4.0
+# Tag: nvidia/cuda:11.8.0-cudnn8-devel-ubuntu20.04
+# Label: com.nvidia.cuda.version: 11.8.0
+# Label: com.nvidia.cudnn.version: 8.7.0
 # Ubuntu 20.04
 FROM nvidia/cuda:11.8.0-cudnn8-devel-ubuntu20.04
 
@@ -11,11 +11,11 @@ ADD scripts /tmp/scripts
 RUN /tmp/scripts/install_ubuntu.sh -p $PYTHON_VERSION && /tmp/scripts/install_os_deps.sh && /tmp/scripts/install_python_deps.sh -p $PYTHON_VERSION && rm -rf /tmp/scripts
 
 # Install TensorRT
-RUN v="8.5.1-1+cuda11.8" &&\
+RUN v="8.6.0.12-1+cuda11.8" &&\
     apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub &&\
     apt-get update &&\
-    sudo apt-get install -y libnvinfer8=${v} libnvonnxparsers8=${v} libnvparsers8=${v} libnvinfer-plugin8=${v} \
-        libnvinfer-dev=${v} libnvonnxparsers-dev=${v} libnvparsers-dev=${v} libnvinfer-plugin-dev=${v} \
+    sudo apt-get install -y libnvinfer8=${v} libnvonnxparsers8=${v} libnvparsers8=${v} libnvinfer-plugin8=${v} libnvinfer-vc-plugin8=${v}\
+        libnvinfer-dev=${v} libnvonnxparsers-dev=${v} libnvparsers-dev=${v} libnvinfer-plugin-dev=${v} libnvinfer-vc-plugin-dev=${v}\
         python3-libnvinfer=${v} libnvinfer-samples=${v}
 
 WORKDIR /root
diff --git a/tools/ci_build/github/linux/run_dockerbuild.sh b/tools/ci_build/github/linux/run_dockerbuild.sh
index 5aa28f0522..c241709fff 100755
--- a/tools/ci_build/github/linux/run_dockerbuild.sh
+++ b/tools/ci_build/github/linux/run_dockerbuild.sh
@@ -96,7 +96,7 @@ elif [ $BUILD_DEVICE = "gpu" ]; then
             --docker-build-args="--build-arg BASEIMAGE=nvcr.io/nvidia/cuda:11.6.2-cudnn8-devel-${BUILD_OS} --build-arg BUILD_USER=onnxruntimedev --build-arg BUILD_UID=$(id -u) --build-arg PYTHON_VERSION=${PYTHON_VER} --build-arg INSTALL_DEPS_EXTRA_ARGS=\"${INSTALL_DEPS_EXTRA_ARGS}\" --build-arg USE_CONDA=${USE_CONDA} --network=host" \
             --dockerfile Dockerfile.ubuntu_gpu_training --context .
 elif [[ $BUILD_DEVICE = "tensorrt"* ]]; then
-        IMAGE="$BUILD_OS-cuda11.8-cudnn8.4-tensorrt8.5"
+        IMAGE="$BUILD_OS-cuda11.8-cudnn8.7-tensorrt8.6"
         DOCKER_FILE=Dockerfile.ubuntu_tensorrt
 
         $GET_DOCKER_IMAGE_CMD --repository "onnxruntime-$IMAGE" \
diff --git a/tools/ci_build/github/windows/setup_env_gpu.bat b/tools/ci_build/github/windows/setup_env_gpu.bat
index 57adb11d90..5f62f42d19 100644
--- a/tools/ci_build/github/windows/setup_env_gpu.bat
+++ b/tools/ci_build/github/windows/setup_env_gpu.bat
@@ -1,6 +1,6 @@
 REM Copyright (c) Microsoft Corporation. All rights reserved.
 REM Licensed under the MIT License.
 
-set PATH=C:\local\TensorRT-8.5.1.7.Windows10.x86_64.cuda-11.8.cudnn8.6\lib;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\extras\CUPTI\lib64;C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\MSBuild\Current\Bin;%PATH%
+set PATH=C:\local\TensorRT-8.6.0.12.Windows10.x86_64.cuda-11.8\lib;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\extras\CUPTI\lib64;C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\MSBuild\Current\Bin;%PATH%
 set GRADLE_OPTS=-Dorg.gradle.daemon=false
 set CUDA_MODULE_LOADING=LAZY