diff --git a/.github/scripts/install-torch-tensorrt.sh b/.github/scripts/install-torch-tensorrt.sh
new file mode 100644
index 0000000000..2930421d5b
--- /dev/null
+++ b/.github/scripts/install-torch-tensorrt.sh
@@ -0,0 +1,9 @@
+#!/usr/bin/env bash
+set -eou pipefail
+# Source conda so it's available to the script environment
+source ${BUILD_ENV_FILE}
+${CONDA_RUN} ${PIP_INSTALL_TORCH} torchvision pyyaml
+export TRT_VERSION=$(${CONDA_RUN} python -c "import versions; versions.tensorrt_version()")
+${CONDA_RUN} python -m pip install /opt/torch-tensorrt-builds/torch_tensorrt*+${CU_VERSION}*.whl tensorrt~=${TRT_VERSION} tensorrt-bindings~=${TRT_VERSION} --extra-index-url=https://pypi.ngc.nvidia.com
+
+echo -e "Running test script";
\ No newline at end of file
diff --git a/.github/scripts/setup-env.sh b/.github/scripts/setup-env.sh
new file mode 100644
index 0000000000..86e814faf5
--- /dev/null
+++ b/.github/scripts/setup-env.sh
@@ -0,0 +1,107 @@
+#!/usr/bin/env bash
+
+set -euxo pipefail
+
+# Prepare conda
+set +x && eval "$($(which conda) shell.bash hook)" && set -x
+
+# Setup the OS_TYPE environment variable that should be used for conditions involving the OS below.
+case $(uname) in
+  Linux)
+    OS_TYPE=linux
+    ;;
+  Darwin)
+    OS_TYPE=macos
+    ;;
+  MSYS*)
+    OS_TYPE=windows
+    ;;
+  *)
+    echo "Unknown OS type:" $(uname)
+    exit 1
+    ;;
+esac
+
+if [[ "${OS_TYPE}" == "macos" && $(uname -m) == x86_64 ]]; then
+  echo '::group::Uninstall system JPEG libraries on macOS'
+  # The x86 macOS runners, e.g. the GitHub Actions native "macos-12" runner, has some JPEG and PNG libraries
+  # installed by default that interfere with our build. We uninstall them here and use the one from conda below.
+  IMAGE_LIBS=$(brew list | grep -E "jpeg|png")
+  for lib in $IMAGE_LIBS; do
+    brew uninstall --ignore-dependencies --force "${lib}"
+  done
+  echo '::endgroup::'
+fi
+
+echo '::group::Create build environment'
+# See https://github.com/pytorch/vision/issues/7296 for ffmpeg
+conda create \
+  --name ci \
+  --quiet --yes \
+  python="${PYTHON_VERSION}" pip \
+  ninja cmake \
+  libpng jpeg \
+  'ffmpeg<4.3'
+conda activate ci
+pip install --progress-bar=off --upgrade setuptools
+
+# See https://github.com/pytorch/vision/issues/6790
+if [[ "${PYTHON_VERSION}" != "3.11" ]]; then
+  pip install --progress-bar=off av!=10.0.0
+fi
+
+echo '::endgroup::'
+
+if [[ "${OS_TYPE}" == windows && "${GPU_ARCH_TYPE}" == cuda ]]; then
+  echo '::group::Install VisualStudio CUDA extensions on Windows'
+  if [[ "${VC_YEAR:-}" == "2022" ]]; then
+    TARGET_DIR="/c/Program Files (x86)/Microsoft Visual Studio/2022/BuildTools/MSBuild/Microsoft/VC/v170/BuildCustomizations"
+  else
+    TARGET_DIR="/c/Program Files (x86)/Microsoft Visual Studio/2019/BuildTools/MSBuild/Microsoft/VC/v160/BuildCustomizations"
+  fi
+  mkdir -p "${TARGET_DIR}"
+  cp -r "${CUDA_HOME}/MSBuildExtensions/"* "${TARGET_DIR}"
+  echo '::endgroup::'
+fi
+
+echo '::group::Install PyTorch'
+# TODO: Can we maybe have this as environment variable in the job template? For example, `IS_RELEASE`.
+if [[ (${GITHUB_EVENT_NAME} = 'pull_request' && (${GITHUB_BASE_REF} = 'release'*)) || (${GITHUB_REF} = 'refs/heads/release'*) ]]; then
+  CHANNEL=test
+else
+  CHANNEL=nightly
+fi
+
+pip install --progress-bar=off light-the-torch
+ltt install --progress-bar=off \
+  --pytorch-computation-backend="${GPU_ARCH_TYPE}${GPU_ARCH_VERSION}" \
+  --pytorch-channel="${CHANNEL}" \
+  torch
+
+if [[ $GPU_ARCH_TYPE == 'cuda' ]]; then
+  python -c "import torch; exit(not torch.cuda.is_available())"
+fi
+echo '::endgroup::'
+
+echo '::group::Install third party dependencies prior to TorchVision install'
+# Installing with `easy_install`, e.g. `python setup.py install` or `python setup.py develop`, has some quirks when
+# when pulling in third-party dependencies. For example:
+# - On Windows, we often hit an SSL error although `pip` can install just fine.
+# - It happily pulls in pre-releases, which can lead to more problems down the line.
+#   `pip` does not unless explicitly told to do so.
+# Thus, we use `easy_install` to extract the third-party dependencies here and install them upfront with `pip`.
+python setup.py egg_info
+# The requires.txt cannot be used with `pip install -r` directly. The requirements are listed at the top and the
+# optional dependencies come in non-standard syntax after a blank line. Thus, we just extract the header.
+sed -e '/^$/,$d' *.egg-info/requires.txt | tee requirements.txt
+pip install --progress-bar=off -r requirements.txt
+echo '::endgroup::'
+
+echo '::group::Install TorchVision'
+python setup.py develop
+echo '::endgroup::'
+
+echo '::group::Collect environment information'
+conda list
+python -m torch.utils.collect_env
+echo '::endgroup::'
\ No newline at end of file
diff --git a/.github/scripts/tests-torchscript-cpp.sh b/.github/scripts/tests-torchscript-cpp.sh
new file mode 100644
index 0000000000..936d2a7c53
--- /dev/null
+++ b/.github/scripts/tests-torchscript-cpp.sh
@@ -0,0 +1,14 @@
+#!/usr/bin/env bash
+
+set -euo pipefail
+
+./.github/scripts/setup-env.sh
+
+# Activate conda environment
+eval "$($(which conda) shell.bash hook)" && conda deactivate && conda activate ci
+
+echo '::group::Install testing utilities'
+pip install --progress-bar=off pytest pytest-mock pytest-cov
+echo '::endgroup::'
+
+pytest --junit-xml="${RUNNER_TEST_RESULTS_DIR}/test-results.xml" -v --durations=25
\ No newline at end of file
diff --git a/.github/workflows/build-test.yml b/.github/workflows/build-test.yml
new file mode 100644
index 0000000000..ae4cd671f0
--- /dev/null
+++ b/.github/workflows/build-test.yml
@@ -0,0 +1,205 @@
+name: Build and test linux wheels
+
+on:
+  pull_request:
+  push:
+    branches:
+      - nightly
+      - main
+      - release/*
+    tags:
+        # NOTE: Binary build pipelines should only get triggered on release candidate builds
+        # Release candidate tags look like: v1.11.0-rc1
+        - v[0-9]+.[0-9]+.[0-9]+-rc[0-9]+
+  workflow_dispatch:
+
+jobs:
+  generate-matrix:
+    uses: pytorch/test-infra/.github/workflows/generate_binary_build_matrix.yml@main
+    with:
+      package-type: wheel
+      os: linux
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      with-rocm: false
+      with-cpu: false
+
+  build:
+    needs: generate-matrix
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - repository: pytorch/tensorrt
+            pre-script: packaging/pre_build_script.sh
+            env-var-script: packaging/env_vars.txt
+            post-script: ""
+            smoke-test-script: ""
+            package-name: torch_tensorrt
+    name: Build torch-tensorrt whl package
+    uses: pytorch/test-infra/.github/workflows/build_wheels_linux.yml@main
+    with:
+      repository: ${{ matrix.repository }}
+      ref: ""
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+      pre-script: ${{ matrix.pre-script }}
+      env-var-script: ${{ matrix.env-var-script }}
+      post-script: ${{ matrix.post-script }}
+      package-name: ${{ matrix.package-name }}
+      smoke-test-script: ${{ matrix.smoke-test-script }}
+      trigger-event: ${{ github.event_name }}
+    secrets:
+      AWS_PYTORCH_UPLOADER_ACCESS_KEY_ID: ${{ secrets.AWS_PYTORCH_UPLOADER_ACCESS_KEY_ID }}
+      AWS_PYTORCH_UPLOADER_SECRET_ACCESS_KEY: ${{ secrets.AWS_PYTORCH_UPLOADER_SECRET_ACCESS_KEY }}
+
+  # tests-py-torchscript-fe:
+  #   name: Test torchscript frontend [Python]
+  #   needs: [generate-matrix, build]
+  #   strategy:
+  #     fail-fast: false
+  #     matrix:
+  #       include:
+  #         - repository: pytorch/tensorrt
+  #           package-name: torch_tensorrt
+  #           pre-script: packaging/pre_build_script.sh
+  #   uses: pytorch/tensorrt/.github/workflows/linux-test.yml@gha-ci-infra
+  #   with:
+  #     job-name: tests-py-torchscript-fe
+  #     repository: "pytorch/tensorrt"
+  #     ref: ""
+  #     test-infra-repository: pytorch/test-infra
+  #     test-infra-ref: main
+  #     build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+  #     pre-script: ${{ matrix.pre-script }}
+  #     script: |
+  #       export USE_HOST_DEPS=1
+  #       pushd .
+  #       cd tests/modules
+  #       ${CONDA_RUN} python -m pip install -r requirements.txt
+  #       ${CONDA_RUN} python hub.py
+  #       popd
+  #       pushd .
+  #       cd tests/py/ts
+  #       ${CONDA_RUN} python -m pip install --pre pytest timm transformers parameterized expecttest --use-deprecated=legacy-resolver
+  #       ${CONDA_RUN} python -m pytest --junitxml=${RUNNER_TEST_RESULTS_DIR}/ts_api_test_results.xml api/
+  #       ${CONDA_RUN} python -m pytest --junitxml=${RUNNER_TEST_RESULTS_DIR}/ts_models_test_results.xml models/
+  #       ${CONDA_RUN} python -m pytest --junitxml=${RUNNER_TEST_RESULTS_DIR}/ts_integrations_test_results.xml integrations/
+  #       popd
+
+  tests-py-dynamo-converters:
+    name: Test dynamo converters [Python]
+    needs: [generate-matrix, build]
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - repository: pytorch/tensorrt
+            package-name: torch_tensorrt
+            pre-script: packaging/pre_build_script.sh
+    uses: pytorch/tensorrt/.github/workflows/linux-test.yml@gha-ci-infra
+    with:
+      job-name: tests-py-dynamo-converters
+      repository: "pytorch/tensorrt"
+      ref: ""
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+      pre-script: ${{ matrix.pre-script }}
+      script: |
+        export USE_HOST_DEPS=1
+        pushd .
+        cd tests/modules
+        ${CONDA_RUN} python -m pip install -r requirements.txt
+        ${CONDA_RUN} python hub.py
+        popd
+        pushd .
+        cd tests/py/dynamo
+        ${CONDA_RUN} python -m pip install --pre pytest-xdist timm transformers parameterized expecttest --use-deprecated=legacy-resolver
+        ${CONDA_RUN} python -m pytest --junitxml=${RUNNER_TEST_RESULTS_DIR}/dynamo_converters_test_results.xml -n 10 conversion/
+        popd
+
+  tests-py-dynamo-fe:
+    name: Test dynamo frontend [Python]
+    needs: [generate-matrix, build]
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - repository: pytorch/tensorrt
+            package-name: torch_tensorrt
+            pre-script: packaging/pre_build_script.sh
+    uses: pytorch/tensorrt/.github/workflows/linux-test.yml@gha-ci-infra
+    with:
+      job-name: tests-py-dynamo-fe
+      repository: "pytorch/tensorrt"
+      ref: ""
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+      pre-script: ${{ matrix.pre-script }}
+      script: |
+        export USE_HOST_DEPS=1
+        pushd .
+        cd tests/py/dynamo
+        ${CONDA_RUN} python -m pip install --pre pytest timm transformers parameterized expecttest --use-deprecated=legacy-resolver
+        ${CONDA_RUN} python -m pytest --junitxml=${RUNNER_TEST_RESULTS_DIR}/dynamo_fe_test_results.xml --ir dynamo models/test_models_export.py
+        popd
+
+  tests-py-torch-compile-be:
+    name: Test torch compile backend [Python]
+    needs: [generate-matrix, build]
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - repository: pytorch/tensorrt
+            package-name: torch_tensorrt
+            pre-script: packaging/pre_build_script.sh
+    uses: pytorch/tensorrt/.github/workflows/linux-test.yml@gha-ci-infra
+    with:
+      job-name: tests-py-torch-compile-be
+      repository: "pytorch/tensorrt"
+      ref: ""
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+      pre-script: ${{ matrix.pre-script }}
+      script: |
+        export USE_HOST_DEPS=1
+        pushd .
+        cd tests/py/dynamo
+        ${CONDA_RUN} python -m pip install --pre pytest-xdist timm transformers parameterized expecttest --use-deprecated=legacy-resolver
+        ${CONDA_RUN} python -m pytest -n 10 --junitxml=${RUNNER_TEST_RESULTS_DIR}/torch_compile_be_test_results.xml backend/
+        ${CONDA_RUN} python -m pytest -n 4 --junitxml=${RUNNER_TEST_RESULTS_DIR}/torch_comple_be_e2e_test_results.xml --ir torch_compile models/test_models.py
+        popd
+
+  tests-py-dynamo-core:
+    name: Test dynamo core [Python]
+    needs: [generate-matrix, build]
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - repository: pytorch/tensorrt
+            package-name: torch_tensorrt
+            pre-script: packaging/pre_build_script.sh
+    uses: pytorch/tensorrt/.github/workflows/linux-test.yml@gha-ci-infra
+    with:
+      job-name: tests-py-dynamo-core
+      repository: "pytorch/tensorrt"
+      ref: ""
+      test-infra-repository: pytorch/test-infra
+      test-infra-ref: main
+      build-matrix: ${{ needs.generate-matrix.outputs.matrix }}
+      pre-script: ${{ matrix.pre-script }}
+      script: |
+        export USE_HOST_DEPS=1
+        pushd .
+        cd tests/py/dynamo
+        ${CONDA_RUN} python -m pip install --pre pytest-xdist timm transformers parameterized expecttest --use-deprecated=legacy-resolver
+        ${CONDA_RUN} python -m pytest -n 10 --junitxml=${RUNNER_TEST_RESULTS_DIR}/tests_py_dynamo_core_runtime_test_results.xml runtime/
+        ${CONDA_RUN} python -m pytest -n 10 --junitxml=${RUNNER_TEST_RESULTS_DIR}/tests_py_dynamo_core_partitioning_test_results.xml partitioning/
+        ${CONDA_RUN} python -m pytest -n 10 --junitxml=${RUNNER_TEST_RESULTS_DIR}/tests_py_dynamo_core_lowering_test_results.xml lowering/
+        popd
\ No newline at end of file
diff --git a/.github/workflows/docgen.yml b/.github/workflows/docgen.yml
index 47dd165d86..eabe615c44 100644
--- a/.github/workflows/docgen.yml
+++ b/.github/workflows/docgen.yml
@@ -10,32 +10,33 @@ on:
 
 jobs:
   build-docs:
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-latest
     container:
       image: docker.io/pytorch/manylinux-builder:cuda12.1
     steps:
       - uses: actions/checkout@v3
         with:
           ref: ${{github.head_ref}}
+      - name: Select python
+        run: |
+          export PATH=/opt/python/cp311-cp311/bin/:$PATH
       - name: Install base deps
         run: |
-          ./packaging/pre_build_script.sh
+          python3 -m pip install pyyaml
+          VERSION_SUFFIX=cu121 ./packaging/pre_build_script.sh
       - name: Get HEAD SHA
         id: vars
         run: echo "sha=$(git rev-parse --short HEAD)" >> $GITHUB_OUTPUT
-      - name: Get Bazel version
-        id: bazel_info
-        run: echo "version=$(cat .bazelversion)" >> $GITHUB_OUTPUT
       - name: Build Python Package
         run: |
-          cp toolchains/ci_workspaces/WORKSPACE.x86_64.cu121.release.rhel WORKSPACE
-          python -m pip install pip<=23
-          python -m pip install --pre -e . --extra-index-url https://download.pytorch.org/whl/nightly/cu121
+          python3 -m pip install pip --upgrade
+          CUDA_HOME=/usr/local/cuda-12.1 CI_BUILD=1 python3 -m pip install --pre -e . --extra-index-url https://download.pytorch.org/whl/nightly/cu121
       - name: Generate New Docs
         run: |
           cd docsrc
-          python -m pip install -r requirements.txt
-          python -c "import torch_tensorrt; print(torch_tensorrt.__version__)"
+          yum install -y doxygen pandoc
+          python3 -m pip install -r requirements.txt
+          python3 -c "import torch_tensorrt; print(torch_tensorrt.__version__)"
           make html
           cd ..
       - uses: stefanzweifel/git-auto-commit-action@v4
diff --git a/.github/workflows/linux-test.yml b/.github/workflows/linux-test.yml
new file mode 100644
index 0000000000..a2b1392f4f
--- /dev/null
+++ b/.github/workflows/linux-test.yml
@@ -0,0 +1,194 @@
+name: Test on Linux
+
+on:
+  workflow_call:
+    inputs:
+      job-name:
+        description: "Name of task"
+        default: ""
+        type: string
+      repository:
+        description: 'Repository to checkout, defaults to ""'
+        default: ""
+        type: string
+      ref:
+        description: 'Reference to checkout, defaults to "nightly"'
+        default: "nightly"
+        type: string
+      pre-script:
+        description: "Pre script to run prior to build"
+        default: ""
+        type: string
+      test-infra-repository:
+        description: "Test infra repository to use"
+        default: "pytorch/test-infra"
+        type: string
+      test-infra-ref:
+        description: "Test infra reference to use"
+        default: ""
+        type: string
+      build-matrix:
+        description: "Build matrix to utilize"
+        default: ""
+        type: string
+      architecture:
+        description: Architecture to build for x86_64 for default Linux, or aarch64 for Linux aarch64 builds
+        required: false
+        type: string
+        default: x86_64
+      setup-miniconda:
+        description: Works as stated in actions/checkout, but the default value is recursive
+        required: false
+        type: boolean
+        default: true
+      script:
+        description: 'Script to utilize'
+        default: "python setup.py bdist_wheel"
+        type: string
+      continue-on-error:
+        description: "Prevents a job from failing when a step fails. Set to true to allow a job to pass when exec script step fails."
+        default: false
+        type: boolean
+      upload-artifact:
+        description: 'Name to give artifacts uploaded from ${RUNNER_ARTIFACT_DIR}'
+        default: ''
+        type: string
+
+jobs:
+  test:
+    strategy:
+      fail-fast: false
+      matrix: ${{ fromJSON(inputs.build-matrix) }}
+    env:
+      PYTHON_VERSION: ${{ matrix.python_version }}
+      PACKAGE_TYPE: wheel
+      REPOSITORY: ${{ inputs.repository }}
+      REF: ${{ inputs.ref }}
+      CU_VERSION: ${{ matrix.desired_cuda }}
+      SCRIPT: ${{ inputs.script }}
+      RUNNER_TEST_RESULTS_DIR: /tmp/test_results
+    name: ${{ inputs.job-name }}-${{ matrix.desired_cuda }}
+    runs-on: ${{ matrix.validation_runner }}
+    container:
+      image: ${{ matrix.container_image }}
+      options: ${{ matrix.gpu_arch_type == 'cuda' && '--gpus all' || ' ' }}
+    # If a build is taking longer than 60 minutes on these runners we need
+    # to have a conversation
+    timeout-minutes: 60
+    steps:
+      - name: Clean workspace
+        run: |
+          set -euxo pipefail
+          echo "::group::Cleanup debug output"
+          rm -rfv "${GITHUB_WORKSPACE}"
+          mkdir -p "${GITHUB_WORKSPACE}"
+          echo "::endgroup::"
+      - uses: actions/checkout@v3
+        with:
+          # Support the use case where we need to checkout someone's fork
+          repository: ${{ inputs.test-infra-repository }}
+          ref: ${{ inputs.test-infra-ref }}
+          path: test-infra
+      - name: Setup SSH
+        uses: ./test-infra/.github/actions/setup-ssh
+        with:
+          github-secret: ${{ github.token }}
+      - uses: ./test-infra/.github/actions/set-channel
+      - uses: ./test-infra/.github/actions/setup-binary-builds
+        with:
+          repository: ${{ inputs.repository }}
+          ref: ${{ inputs.ref }}
+          setup-miniconda: ${{ inputs.setup-miniconda }}
+          python-version: ${{ env.PYTHON_VERSION }}
+      - name: Run Pre-Script with Caching
+        if: ${{ inputs.pre-script != '' }}
+        uses: ./test-infra/.github/actions/run-script-with-cache
+        with:
+          cache-path: ${{ inputs.cache-path }}
+          cache-key: ${{ inputs.cache-key }}
+          repository: ${{ inputs.repository  }}
+          script: ${{ inputs.pre-script }}
+      - name: Download artifacts
+        uses: actions/download-artifact@v3
+        with:
+          name: ${{ env.ARTIFACT_NAME }}
+          path: /opt/torch-tensorrt-builds/
+      # - name: Install torch and torch-tensorrt
+      #   if: ${{ inputs.pre-script != '' }}
+      #   uses: ./test-infra/.github/actions/run-script-with-cache
+      #   with:
+      #     repository: ${{ inputs.repository  }}
+      #     script: .github/scripts/install-torch-tensorrt.sh
+      - name: Pack script
+        continue-on-error: ${{ inputs.continue-on-error }}
+        working-directory: ${{ inputs.repository }}
+        env:
+          ALL_SECRETS: ${{ toJSON(secrets) }}
+        run: |
+          set -euxo pipefail
+          # shellcheck disable=SC2086
+          source "${BUILD_ENV_FILE}"
+          {
+            echo "${SCRIPT}";
+          } > "user_script"
+          cat .github/scripts/install-torch-tensorrt.sh user_script > exec_script
+      - name: Run Script
+        uses: ./test-infra/.github/actions/run-script-with-cache
+        with:
+          repository: ${{ inputs.repository  }}
+          script: exec_script
+      - name: Surface failing tests
+        if: always()
+        uses: pmeier/pytest-results-action@v0.3.0
+        with:
+          path: ${{ env.RUNNER_TEST_RESULTS_DIR }}
+          fail-on-empty: false
+
+      - name: Prepare artifacts for upload
+        working-directory: ${{ inputs.repository }}
+        id: check-artifacts
+        env:
+          UPLOAD_ARTIFACT_NAME: ${{ inputs.upload-artifact }}
+        run: |
+          # Only do these steps if we actually want to upload an artifact
+          if [[ -n "${UPLOAD_ARTIFACT_NAME}" ]]; then
+            # If the default execution path is followed then we should get a wheel in the dist/ folder
+            # attempt to just grab whatever is in there and scoop it all up
+            if find "dist/" -name "*.whl" >/dev/null 2>/dev/null; then
+              mv -v dist/*.whl "${RUNNER_ARTIFACT_DIR}/"
+            fi
+            # Set to fail upload step if there are no files for upload and expected files for upload
+            echo 'if-no-files-found=error' >> "${GITHUB_OUTPUT}"
+          fi
+
+          upload_docs=0
+          # Check if there are things in the documentation folder to uplaod
+          if find "${RUNNER_DOCS_DIR}" -mindepth 1 -maxdepth 1 | read -r; then
+            # TODO: Add a check here to test if on ec2 because if we're not on ec2 then this
+            # upload will probably not work correctly
+            upload_docs=1
+          fi
+          echo "upload-docs=${upload_docs}" >> "${GITHUB_OUTPUT}"
+
+      - name: Upload artifacts to GitHub (if any)
+        uses: actions/upload-artifact@v3
+        if: ${{ inputs.upload-artifact != '' }}
+        with:
+          name: ${{ inputs.upload-artifact }}
+          path: ${{ runner.temp }}/artifacts/
+          if-no-files-found: ${{ steps.check-artifacts.outputs.if-no-files-found }}
+
+      - name: Upload documentation to S3 (if any)
+        uses: seemethere/upload-artifact-s3@v5
+        if: ${{ steps.check-artifacts.outputs.upload-docs == 1 && github.event.pull_request.number != '' }}
+        with:
+          retention-days: 14
+          s3-bucket: doc-previews
+          if-no-files-found: error
+          path: ${{ env.RUNNER_DOCS_DIR }}
+          # ${{ env.repository }} is $OWNER/$REPO
+          s3-prefix: ${{ env.REPOSITORY }}/${{ github.event.pull_request.number }}
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.sha }}-${{ inputs.repository }}-${{ github.event_name == 'workflow_dispatch' }}-${{ inputs.job-name }}
+  cancel-in-progress: true
\ No newline at end of file
diff --git a/.github/workflows/nightlies.yml b/.github/workflows/nightlies.yml
new file mode 100644
index 0000000000..31e3b034f6
--- /dev/null
+++ b/.github/workflows/nightlies.yml
@@ -0,0 +1,27 @@
+name: Cut Nightly
+
+on:
+  schedule:
+    - cron: '30 9 * * *'
+  workflow_dispatch:
+
+jobs:
+  cut_nightly:
+    runs-on: ubuntu-latest
+    timeout-minutes: 60
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          ref: main
+      - name: Create nightly commit
+        shell: bash
+        run: |
+          git config --global user.email "torch-tensorrt.github.bot@nvidia.com"
+          git config --global user.name "Torch-TensorRT Github Bot"
+          git fetch origin nightly
+          HEAD_COMMIT_HASH=$(git rev-parse HEAD)
+          NIGHTLY_DATE=$(date +"%Y-%m-%d")
+          # shellcheck disable=SC1083
+          NIGHTLY_RELEASE_COMMIT=$(git commit-tree -p FETCH_HEAD HEAD^{tree} -m "${NIGHTLY_DATE} nightly release (${HEAD_COMMIT_HASH})")
+          # shellcheck disable=SC1083
+          git push -f origin "${NIGHTLY_RELEASE_COMMIT}:nightly"
\ No newline at end of file
diff --git a/WORKSPACE b/WORKSPACE
index ae98bc949c..b24384dfe8 100644
--- a/WORKSPACE
+++ b/WORKSPACE
@@ -53,17 +53,15 @@ new_local_repository(
 http_archive(
     name = "libtorch",
     build_file = "@//third_party/libtorch:BUILD",
-    sha256 = "1ae8366aaf7af7f68f142ba644fe26c837c6fa8347ec6bd9ce605ac60e7f7e5e",
     strip_prefix = "libtorch",
-    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-2.1.0.dev20230703%2Bcu121.zip"],
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-latest.zip"],
 )
 
 http_archive(
     name = "libtorch_pre_cxx11_abi",
     build_file = "@//third_party/libtorch:BUILD",
-    sha256 = "9add4832f4da9223866d85810820b816ab3319d5a227066101eeb6cbb76adb4b",
     strip_prefix = "libtorch",
-    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-2.1.0.dev20230703%2Bcu121.zip"],
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-latest.zip"],
 )
 
 # Download these tarballs manually from the NVIDIA website
diff --git a/core/runtime/TRTEngine.h b/core/runtime/TRTEngine.h
index 1a2df1f2e3..3f165ee0c0 100644
--- a/core/runtime/TRTEngine.h
+++ b/core/runtime/TRTEngine.h
@@ -1,5 +1,6 @@
 #pragma once
 #include <experimental/filesystem>
+#include <fstream>
 #include <map>
 #include <memory>
 #include <mutex>
diff --git a/versions.yml b/dev_dep_versions.yml
similarity index 100%
rename from versions.yml
rename to dev_dep_versions.yml
diff --git a/packaging/env_vars.txt b/packaging/env_vars.txt
new file mode 100644
index 0000000000..46f906b1ff
--- /dev/null
+++ b/packaging/env_vars.txt
@@ -0,0 +1,2 @@
+export CI_BUILD="1"
+export RELEASE="1"
\ No newline at end of file
diff --git a/packaging/post_build_script.sh b/packaging/post_build_script.sh
new file mode 100644
index 0000000000..e69de29bb2
diff --git a/packaging/pre_build_script.sh b/packaging/pre_build_script.sh
old mode 100644
new mode 100755
index 49a3a187d5..18cd5d9fe2
--- a/packaging/pre_build_script.sh
+++ b/packaging/pre_build_script.sh
@@ -1,13 +1,16 @@
 #!/bin/bash
 
 # Install dependencies
+python3 -m pip install pyyaml
 TRT_VERSION=$(python3 -c "import versions; versions.tensorrt_version()")
 yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo
 yum check-update
-yum install -y ninja-build tensorrt-${TRT_VERSION}.*
+yum install -y ninja-build gettext tensorrt-${TRT_VERSION}.*
 wget https://github.com/bazelbuild/bazelisk/releases/download/v1.17.0/bazelisk-linux-amd64 \
     && mv bazelisk-linux-amd64 /usr/bin/bazel \
     && chmod +x /usr/bin/bazel
 
-cp toolchains/ci_workspaces/WORKSPACE.x86_64.${VERSION_SUFFIX#*+}.release.rhel WORKSPACE
+export TORCH_BUILD_NUMBER=$(python -c "import torch, urllib.parse as ul; print(ul.quote_plus(torch.__version__))")
+
+cat toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel.tmpl | envsubst > WORKSPACE
 export CI_BUILD=1
diff --git a/py/ci/build_whl.sh b/py/ci/build_whl.sh
index 0998e87384..12f72a9351 100755
--- a/py/ci/build_whl.sh
+++ b/py/ci/build_whl.sh
@@ -6,11 +6,23 @@ export CXX=g++
 export CUDA_HOME=/usr/local/cuda-12.1
 export PROJECT_DIR=/workspace/project
 
-cp -r $CUDA_HOME /usr/local/cuda
+rm -rf /usr/local/cuda
+
+if [[ $CUDA_HOME == "/usr/local/cuda-12.1" ]]; then
+    cp -r /usr/local/cuda-11.8 /usr/local/cuda
+    cp -r /usr/local/cuda-12.0/ /usr/local/cuda/
+    rsync -a /usr/local/cuda-12.1/ /usr/local/cuda/
+    export CUDA_HOME=/usr/local/cuda
+else
+    ln -s $CUDA_HOME /usr/local/cuda
+fi
 
 build_wheel() {
-    $1/bin/python -m pip install --upgrade pip
-    $1/bin/python -m pip wheel . --config-setting="--build-option=--release" --config-setting="--build-option=--ci" -w dist
+    $1/bin/python -m pip install --upgrade pip setuptools
+    $1/bin/python -m pip install -r py/requirements.txt
+    #$1/bin/python -m pip wheel . -w dist
+    export BUILD_VERSION=$(cd ${PROJECT_DIR} && $1/bin/python3 -c "import versions; versions.torch_tensorrt_version_release()")
+    CI_BUILD=1 RELEASE=1 $1/bin/python setup.py bdist_wheel
 }
 
 patch_wheel() {
@@ -86,7 +98,7 @@ libtorchtrt() {
     ${PY_DIR}/bin/python -m pip install setuptools wheel auditwheel
     bazel build //:libtorchtrt --platforms //toolchains:ci_rhel_x86_64_linux -c opt --noshow_progress
     CUDA_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.cuda_version()")
-    TORCHTRT_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.torch_tensorrt_version()")
+    TORCHTRT_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.torch_tensorrt_version_release()")
     TRT_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.tensorrt_version()")
     CUDNN_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.cudnn_version()")
     TORCH_VERSION=$(${PY_DIR}/bin/python -c "from torch import __version__;print(__version__.split('+')[0])")
@@ -106,7 +118,7 @@ libtorchtrt_pre_cxx11_abi() {
     ${PY_DIR}/bin/python -m pip install setuptools wheel auditwheel
     bazel build //:libtorchtrt --config pre_cxx11_abi --platforms //toolchains:ci_rhel_x86_64_linux -c opt --noshow_progress
     CUDA_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.cuda_version()")
-    TORCHTRT_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.torch_tensorrt_version()")
+    TORCHTRT_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.torch_tensorrt_version_release()")
     TRT_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.tensorrt_version()")
     CUDNN_VERSION=$(cd ${PROJECT_DIR} && ${PY_DIR}/bin/python3 -c "import versions; versions.cudnn_version()")
     TORCH_VERSION=$(${PY_DIR}/bin/python -c "from torch import __version__;print(__version__.split('+')[0])")
diff --git a/py/requirements.txt b/py/requirements.txt
index 5c91e3684e..0116b23c4f 100644
--- a/py/requirements.txt
+++ b/py/requirements.txt
@@ -2,8 +2,8 @@ numpy
 packaging
 pybind11==2.6.2
 --extra-index-url https://download.pytorch.org/whl/nightly/cu121
-torch==2.1.0.dev20230703+cu121
-torchvision==0.16.0.dev20230703+cu121
+torch>=2.1.0.dev,<2.2.0
+torchvision>=0.16.0.dev,<0.17.0
 --extra-index-url https://pypi.ngc.nvidia.com
 tensorrt==8.6.1
 pyyaml
\ No newline at end of file
diff --git a/py/torch_tensorrt/__init__.py b/py/torch_tensorrt/__init__.py
index 11c5cc11ec..c24ef3897f 100644
--- a/py/torch_tensorrt/__init__.py
+++ b/py/torch_tensorrt/__init__.py
@@ -59,9 +59,7 @@ def _find_lib(name: str, paths: List[str]) -> str:
             ctypes.CDLL(_find_lib(lib, WIN_PATHS))
 
     elif sys.platform.startswith("linux"):
-        LINUX_PATHS = [
-            "/usr/local/cuda-12.1/lib64",
-        ]
+        LINUX_PATHS = ["/usr/local/cuda-12.1/lib64", "/usr/lib", "/usr/lib64"]
 
         if "LD_LIBRARY_PATH" in os.environ:
             LINUX_PATHS += os.environ["LD_LIBRARY_PATH"].split(os.path.pathsep)
diff --git a/py/torch_tensorrt/_compile.py b/py/torch_tensorrt/_compile.py
index 7106a4e6be..08e7f7d424 100644
--- a/py/torch_tensorrt/_compile.py
+++ b/py/torch_tensorrt/_compile.py
@@ -35,13 +35,13 @@
 def _non_fx_input_interface(
     inputs: Sequence[Input | torch.Tensor | InputTensorSpec],
 ) -> TypeGuard[List[Input | torch.Tensor]]:
-    return all(isinstance(i, torch.Tensor | Input) for i in inputs)
+    return all(isinstance(i, (torch.Tensor, Input)) for i in inputs)
 
 
 def _fx_input_interface(
     inputs: Sequence[Input | torch.Tensor | InputTensorSpec],
 ) -> TypeGuard[List[InputTensorSpec | torch.Tensor]]:
-    return all(isinstance(i, torch.Tensor | InputTensorSpec) for i in inputs)
+    return all(isinstance(i, (torch.Tensor, InputTensorSpec)) for i in inputs)
 
 
 class _IRType(Enum):
diff --git a/pyproject.toml b/pyproject.toml
index 3f99213db9..3973f1fdb9 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -9,8 +9,8 @@ requires = [
     "typing-extensions>=4.7.0",
     "future>=0.18.3",
     "tensorrt>=8.6,<8.7",
-    #"torch >=2.0.0,<2.1.0",
-    "torch==2.1.0.dev20230703+cu121",
+    "torch >=2.1.0.dev,<2.2.0",
+    #"torch==2.1.0.dev20230731",
     "pybind11==2.6.2",
     "numpy",
 ]
@@ -42,8 +42,8 @@ readme = {file = "py/README.md", content-type = "text/markdown"}
 requires-python = ">=3.8"
 keywords = ["pytorch", "torch", "tensorrt", "trt", "ai", "artificial intelligence", "ml", "machine learning", "dl", "deep learning", "compiler", "dynamo", "torchscript", "inference"]
 dependencies = [
-    #"torch>=2.0.0,<2.1.0",
-    "torch==2.1.0.dev20230703+cu121",
+    "torch >=2.1.0.dev,<2.2.0",
+    #"torch==2.1.0.dev20230731",
     "tensorrt>=8.6,<8.7",
     "packaging>=23",
     "numpy",
diff --git a/setup.py b/setup.py
index 0964cc53fe..fec696bbce 100644
--- a/setup.py
+++ b/setup.py
@@ -18,35 +18,29 @@
 from torch.utils import cpp_extension
 from wheel.bdist_wheel import bdist_wheel
 
-dir_path = os.path.dirname(os.path.realpath(__file__)) + "/py"
+import yaml
+import re
+import os
+import subprocess
 
-CXX11_ABI = False
-JETPACK_VERSION = None
-FX_ONLY = False
-LEGACY = False
-RELEASE = False
-CI_RELEASE = False
+from datetime import datetime
+from pathlib import Path
+from typing import List
 
 __version__: str = "0.0.0"
 __cuda_version__: str = "0.0"
 __cudnn_version__: str = "0.0"
 __tensorrt_version__: str = "0.0"
 
-
-def load_version_info():
-    global __version__
-    global __cuda_version__
-    global __cudnn_version__
-    global __tensorrt_version__
-    with open("versions.yml", "r") as stream:
-        versions = yaml.safe_load(stream)
-        __version__ = versions["__version__"]
-        __cuda_version__ = versions["__cuda_version__"]
-        __cudnn_version__ = versions["__cudnn_version__"]
-        __tensorrt_version__ = versions["__tensorrt_version__"]
+LEGACY_BASE_VERSION_SUFFIX_PATTERN = re.compile("a0$")
 
 
-load_version_info()
+def get_root_dir() -> Path:
+    return Path(
+        subprocess.check_output(["git", "rev-parse", "--show-toplevel"])
+        .decode("ascii")
+        .strip()
+    )
 
 
 def get_git_revision_short_hash() -> str:
@@ -57,6 +51,45 @@ def get_git_revision_short_hash() -> str:
     )
 
 
+def get_base_version() -> str:
+    root = get_root_dir()
+    try:
+        dirty_version = open(root / "version.txt", "r").read().strip()
+    except FileNotFoundError:
+        print("# WARNING: Base version not found defaulting BUILD_VERSION to 0.1.0")
+        dirty_version = "0.1.0"
+    # Strips trailing a0 from version.txt, not too sure why it's there in the
+    # first place
+    return re.sub(LEGACY_BASE_VERSION_SUFFIX_PATTERN, "", dirty_version)
+
+
+def load_dep_info():
+    global __cuda_version__
+    global __cudnn_version__
+    global __tensorrt_version__
+    with open("dev_dep_versions.yml", "r") as stream:
+        versions = yaml.safe_load(stream)
+        if (gpu_arch_version := os.environ.get("CU_VERSION")) is not None:
+            __cuda_version__ = (
+                (gpu_arch_version[2:])[:-1] + "." + (gpu_arch_version[2:])[-1:]
+            )
+        else:
+            __cuda_version__ = versions["__cuda_version__"]
+        __cudnn_version__ = versions["__cudnn_version__"]
+        __tensorrt_version__ = versions["__tensorrt_version__"]
+
+
+load_dep_info()
+
+dir_path = str(get_root_dir()) + "/py"
+
+CXX11_ABI = False
+JETPACK_VERSION = None
+FX_ONLY = False
+LEGACY = False
+RELEASE = False
+CI_BUILD = False
+
 if "--fx-only" in sys.argv:
     FX_ONLY = True
     sys.argv.remove("--fx-only")
@@ -65,16 +98,31 @@ def get_git_revision_short_hash() -> str:
     LEGACY = True
     sys.argv.remove("--legacy")
 
-if "--release" not in sys.argv:
-    __version__ = __version__ + "+" + get_git_revision_short_hash()
-else:
+if "--release" in sys.argv:
     RELEASE = True
     sys.argv.remove("--release")
 
+if (release_env_var := os.environ.get("RELEASE")) is not None:
+    if release_env_var == "1":
+        RELEASE = True
+
+if (gpu_arch_version := os.environ.get("CU_VERSION")) is None:
+    gpu_arch_version = f"cu{__cuda_version__.replace('.','')}"
+
+
+if RELEASE:
+    __version__ = os.environ.get("BUILD_VERSION")
+else:
+    __version__ = f"{get_base_version()}.dev0+{get_git_revision_short_hash()}"
+
 if "--ci" in sys.argv:
     sys.argv.remove("--ci")
     if RELEASE:
-        CI_RELEASE = True
+        CI_BUILD = True
+
+if (ci_env_var := os.environ.get("CI_BUILD")) is not None:
+    if ci_env_var == "1":
+        CI_BUILD = True
 
 if "--use-cxx11-abi" in sys.argv:
     sys.argv.remove("--use-cxx11-abi")
@@ -158,7 +206,7 @@ def build_libtorchtrt_pre_cxx11_abi(develop=True, use_dist_dir=True, cxx11_abi=F
         cmd.append("--platforms=//toolchains:jetpack_5.0")
         print("Jetpack version: 5.0")
 
-    if CI_RELEASE:
+    if CI_BUILD:
         cmd.append("--platforms=//toolchains:ci_rhel_x86_64_linux")
         print("CI based build")
 
@@ -415,6 +463,7 @@ def run(self):
                 dir_path + "/../bazel-TensorRT/external/tensorrt/include",
                 dir_path + "/../bazel-tensorrt/external/tensorrt/include",
                 dir_path + "/../",
+                "/usr/local/cuda",
             ],
             extra_compile_args=[
                 "-Wno-deprecated",
diff --git a/toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel b/toolchains/ci_workspaces/WORKSPACE.x86_64.cu118.release.rhel
similarity index 91%
rename from toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel
rename to toolchains/ci_workspaces/WORKSPACE.x86_64.cu118.release.rhel
index 8d6ff234e8..79c7575d2b 100644
--- a/toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel
+++ b/toolchains/ci_workspaces/WORKSPACE.x86_64.cu118.release.rhel
@@ -43,7 +43,7 @@ local_repository(
 new_local_repository(
     name = "cuda",
     build_file = "@//third_party/cuda:BUILD",
-    path = "/usr/local/cuda-12.1",
+    path = "/usr/local/cuda-11.8",
 )
 
 new_local_repository(
@@ -58,17 +58,17 @@ new_local_repository(
 http_archive(
     name = "libtorch",
     build_file = "@//third_party/libtorch:BUILD",
-    sha256 = "1ae8366aaf7af7f68f142ba644fe26c837c6fa8347ec6bd9ce605ac60e7f7e5e",
+    sha256 = "174579a7ee2a506d063714160c5fc57da428f7935311ef511c8f19820eb14c86",
     strip_prefix = "libtorch",
-    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-2.1.0.dev20230703%2Bcu121.zip"],
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-2.1.0.dev20230731%2Bcu121.zip"],
 )
 
 http_archive(
     name = "libtorch_pre_cxx11_abi",
     build_file = "@//third_party/libtorch:BUILD",
-    sha256 = "9add4832f4da9223866d85810820b816ab3319d5a227066101eeb6cbb76adb4b",
+    sha256 = "532217063c65354d5534211badadc9c370d889cb1c3fdb295c9b3d0f181bc0ba",
     strip_prefix = "libtorch",
-    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-2.1.0.dev20230703%2Bcu121.zip"],
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-2.1.0.dev20230731%2Bcu121.zip"],
 )
 
 ####################################################################################
diff --git a/toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel.tmpl b/toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel.tmpl
new file mode 100644
index 0000000000..00b85034fe
--- /dev/null
+++ b/toolchains/ci_workspaces/WORKSPACE.x86_64.release.rhel.tmpl
@@ -0,0 +1,101 @@
+workspace(name = "Torch-TensorRT")
+
+load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")
+
+http_archive(
+    name = "rules_python",
+    sha256 = "863ba0fa944319f7e3d695711427d9ad80ba92c6edd0b7c7443b84e904689539",
+    strip_prefix = "rules_python-0.22.0",
+    url = "https://github.com/bazelbuild/rules_python/releases/download/0.22.0/rules_python-0.22.0.tar.gz",
+)
+
+load("@rules_python//python:repositories.bzl", "py_repositories")
+
+py_repositories()
+
+http_archive(
+    name = "rules_pkg",
+    sha256 = "8f9ee2dc10c1ae514ee599a8b42ed99fa262b757058f65ad3c384289ff70c4b8",
+    urls = [
+        "https://mirror.bazel.build/github.com/bazelbuild/rules_pkg/releases/download/0.9.1/rules_pkg-0.9.1.tar.gz",
+        "https://github.com/bazelbuild/rules_pkg/releases/download/0.9.1/rules_pkg-0.9.1.tar.gz",
+    ],
+)
+
+load("@rules_pkg//:deps.bzl", "rules_pkg_dependencies")
+
+rules_pkg_dependencies()
+
+http_archive(
+    name = "googletest",
+    sha256 = "755f9a39bc7205f5a0c428e920ddad092c33c8a1b46997def3f1d4a82aded6e1",
+    strip_prefix = "googletest-5ab508a01f9eb089207ee87fd547d290da39d015",
+    urls = ["https://github.com/google/googletest/archive/5ab508a01f9eb089207ee87fd547d290da39d015.zip"],
+)
+
+# External dependency for torch_tensorrt if you already have precompiled binaries.
+local_repository(
+    name = "torch_tensorrt",
+    path = "/opt/circleci/.pyenv/versions/3.10.9/lib/python3.10/site-packages/torch_tensorrt"
+)
+
+# CUDA should be installed on the system locally
+new_local_repository(
+    name = "cuda",
+    build_file = "@//third_party/cuda:BUILD",
+    path = "${CUDA_HOME}",
+)
+
+new_local_repository(
+    name = "cublas",
+    build_file = "@//third_party/cublas:BUILD",
+    path = "/usr",
+)
+#############################################################################################################
+# Tarballs and fetched dependencies (default - use in cases when building from precompiled bin and tarballs)
+#############################################################################################################
+
+http_archive(
+    name = "libtorch",
+    build_file = "@//third_party/libtorch:BUILD",
+    strip_prefix = "libtorch",
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-latest.zip"],
+)
+
+http_archive(
+    name = "libtorch_pre_cxx11_abi",
+    build_file = "@//third_party/libtorch:BUILD",
+    strip_prefix = "libtorch",
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-latest.zip"],
+)
+
+####################################################################################
+# Locally installed dependencies (use in cases of custom dependencies or aarch64)
+####################################################################################
+
+new_local_repository(
+    name = "cudnn",
+    path = "/usr/",
+    build_file = "@//third_party/cudnn/local:BUILD"
+)
+
+new_local_repository(
+   name = "tensorrt",
+   path = "/usr/",
+   build_file = "@//third_party/tensorrt/local:BUILD"
+)
+
+# #########################################################################
+# # Testing Dependencies (optional - comment out on aarch64)
+# #########################################################################
+
+load("@rules_python//python:pip.bzl", "pip_parse")
+
+pip_parse(
+    name = "devtools_deps",
+    requirements = "//:requirements-dev.txt",
+)
+
+load("@devtools_deps//:requirements.bzl", "install_deps")
+
+install_deps()
diff --git a/toolchains/ci_workspaces/WORKSPACE.x86_64.release.ubuntu b/toolchains/ci_workspaces/WORKSPACE.x86_64.release.ubuntu
index 60e29f2b60..745272d328 100644
--- a/toolchains/ci_workspaces/WORKSPACE.x86_64.release.ubuntu
+++ b/toolchains/ci_workspaces/WORKSPACE.x86_64.release.ubuntu
@@ -58,17 +58,17 @@ new_local_repository(
 http_archive(
     name = "libtorch",
     build_file = "@//third_party/libtorch:BUILD",
-    sha256 = "1ae8366aaf7af7f68f142ba644fe26c837c6fa8347ec6bd9ce605ac60e7f7e5e",
+    sha256 = "174579a7ee2a506d063714160c5fc57da428f7935311ef511c8f19820eb14c86",
     strip_prefix = "libtorch",
-    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-2.1.0.dev20230703%2Bcu121.zip"],
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-cxx11-abi-shared-with-deps-2.1.0.dev20230731%2Bcu121.zip"],
 )
 
 http_archive(
     name = "libtorch_pre_cxx11_abi",
     build_file = "@//third_party/libtorch:BUILD",
-    sha256 = "9add4832f4da9223866d85810820b816ab3319d5a227066101eeb6cbb76adb4b",
+    sha256 = "532217063c65354d5534211badadc9c370d889cb1c3fdb295c9b3d0f181bc0ba",
     strip_prefix = "libtorch",
-    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-2.1.0.dev20230703%2Bcu121.zip"],
+    urls = ["https://download.pytorch.org/libtorch/nightly/cu121/libtorch-shared-with-deps-2.1.0.dev20230731%2Bcu121.zip"],
 )
 
 ####################################################################################
diff --git a/version.txt b/version.txt
new file mode 100644
index 0000000000..32408126ec
--- /dev/null
+++ b/version.txt
@@ -0,0 +1 @@
+2.0.0a0
\ No newline at end of file
diff --git a/versions.py b/versions.py
index f038aabb24..772737aab7 100644
--- a/versions.py
+++ b/versions.py
@@ -1,29 +1,141 @@
 import yaml
+import re
+import os
+import subprocess
 
-__version__: str = "0.0.0"
-__cuda_version__: str = "0.0"
-__cudnn_version__: str = "0.0"
-__tensorrt_version__: str = "0.0"
+from datetime import datetime
+from pathlib import Path
+from typing import List
 
+__version__ = "0.0.0"
+__cuda_version__ = "0.0"
+__cudnn_version__ = "0.0"
+__tensorrt_version__ = "0.0"
 
-def load_version_info():
-    global __version__
+
+LEADING_V_PATTERN = re.compile("^v")
+TRAILING_RC_PATTERN = re.compile("-rc[0-9]*$")
+LEGACY_BASE_VERSION_SUFFIX_PATTERN = re.compile("a0$")
+
+
+class NoGitTagException(Exception):
+    pass
+
+
+def get_root_dir() -> Path:
+    return Path(
+        subprocess.check_output(["git", "rev-parse", "--show-toplevel"])
+        .decode("ascii")
+        .strip()
+    )
+
+
+def get_tag() -> str:
+    root = get_root_dir()
+    # We're on a tag
+    am_on_tag = (
+        subprocess.run(
+            ["git", "describe", "--tags", "--exact"],
+            cwd=root,
+            stdout=subprocess.DEVNULL,
+            stderr=subprocess.DEVNULL,
+        ).returncode
+        == 0
+    )
+    tag = ""
+    if am_on_tag:
+        dirty_tag = (
+            subprocess.check_output(["git", "describe"], cwd=root)
+            .decode("ascii")
+            .strip()
+        )
+        # Strip leading v that we typically do when we tag branches
+        # ie: v1.7.1 -> 1.7.1
+        tag = re.sub(LEADING_V_PATTERN, "", dirty_tag)
+        # Strip trailing rc pattern
+        # ie: 1.7.1-rc1 -> 1.7.1
+        tag = re.sub(TRAILING_RC_PATTERN, "", tag)
+    return tag
+
+
+def get_base_version() -> str:
+    root = get_root_dir()
+    try:
+        dirty_version = open(root / "version.txt", "r").read().strip()
+    except FileNotFoundError:
+        print("# WARNING: Base version not found defaulting BUILD_VERSION to 0.1.0")
+        dirty_version = "0.1.0"
+    # Strips trailing a0 from version.txt, not too sure why it's there in the
+    # first place
+    return re.sub(LEGACY_BASE_VERSION_SUFFIX_PATTERN, "", dirty_version)
+
+
+class PytorchVersion:
+    def __init__(
+        self,
+        gpu_arch_version: str,
+        no_build_suffix: bool,
+        base_build_version: str,
+    ) -> None:
+        self.gpu_arch_version = gpu_arch_version
+        self.no_build_suffix = no_build_suffix
+        if base_build_version == "":
+            base_build_version = get_base_version()
+        self.base_build_version = base_build_version
+
+    def get_post_build_suffix(self) -> str:
+        if self.no_build_suffix or self.gpu_arch_version is None:
+            return ""
+        return f"+{self.gpu_arch_version}"
+
+    def get_release_version(self) -> str:
+        if self.base_build_version:
+            return f"{self.base_build_version}{self.get_post_build_suffix()}"
+        if not get_tag():
+            raise NoGitTagException(
+                "Not on a git tag, are you sure you want a release version?"
+            )
+        return f"{get_tag()}{self.get_post_build_suffix()}"
+
+    def get_nightly_version(self) -> str:
+        date_str = datetime.today().strftime("%Y%m%d")
+        build_suffix = self.get_post_build_suffix()
+        return f"{self.base_build_version}.dev{date_str}{build_suffix}"
+
+
+def load_dep_info():
     global __cuda_version__
     global __cudnn_version__
     global __tensorrt_version__
-    with open("versions.yml", "r") as stream:
+    with open("dev_dep_versions.yml", "r") as stream:
         versions = yaml.safe_load(stream)
-        __version__ = versions["__version__"]
-        __cuda_version__ = versions["__cuda_version__"]
+        gpu_arch_version = os.environ.get("CU_VERSION")
+        if gpu_arch_version is not None:
+            __cuda_version__ = (
+                (gpu_arch_version[2:])[:-1] + "." + (gpu_arch_version[2:])[-1:]
+            )
+        else:
+            __cuda_version__ = versions["__cuda_version__"]
         __cudnn_version__ = versions["__cudnn_version__"]
         __tensorrt_version__ = versions["__tensorrt_version__"]
 
 
-load_version_info()
+load_dep_info()
+
+gpu_arch_version = os.environ.get("CU_VERSION")
+version = PytorchVersion(
+    gpu_arch_version=gpu_arch_version,
+    no_build_suffix=False,
+    base_build_version=get_base_version(),
+)
+
+
+def torch_tensorrt_version_nightly():
+    print(version.get_nightly_version())
 
 
-def torch_tensorrt_version():
-    print(__version__)
+def torch_tensorrt_version_release():
+    print(version.get_release_version())
 
 
 def cuda_version():