diff --git a/README.md b/README.md
index 0a7e8bddb26..70accc10a3c 100644
--- a/README.md
+++ b/README.md
@@ -60,11 +60,17 @@ Learn more: [How ExecuTorch Works](https://docs.pytorch.org/executorch/main/intr
 
 ### Installation
 
+**Python (via pip):**
 ```bash
 pip install executorch
 ```
 
-For platform-specific setup (Android, iOS, embedded systems), see the [Quick Start](https://docs.pytorch.org/executorch/main/quick-start-section.html) documentation for additional info.
+**C++ (via VCPKG):**
+```bash
+vcpkg install executorch
+```
+
+For platform-specific setup (Android, iOS, embedded systems), see the [Quick Start](https://docs.pytorch.org/executorch/main/quick-start-section.html) documentation for additional info. For VCPKG port details and features, see the [ports/README.md](ports/README.md).
 
 ### Export and Deploy in 3 Steps
 
diff --git a/ports/README.md b/ports/README.md
new file mode 100644
index 00000000000..42db7791c2a
--- /dev/null
+++ b/ports/README.md
@@ -0,0 +1,69 @@
+# ExecuTorch VCPKG Port
+
+This directory contains the VCPKG port configuration for ExecuTorch, PyTorch's on-device AI inference framework.
+
+## About VCPKG Ports
+
+VCPKG is a C/C++ package manager from Microsoft. A "port" is a set of scripts and metadata that tells VCPKG how to download, build, and install a library.
+
+## Files in this Directory
+
+- **vcpkg.json**: Manifest file containing package metadata, version, dependencies, and optional features
+- **portfile.cmake**: Main build script that tells VCPKG how to configure, build, and install ExecuTorch
+- **usage**: Instructions for users on how to use the installed package in their CMake projects
+
+## Using this Port
+
+### For VCPKG Registry Maintainers
+
+To submit this port to the official VCPKG registry:
+
+1. Follow the VCPKG contribution guidelines: https://github.com/microsoft/vcpkg/blob/master/docs/maintainers/control-files.md
+2. Copy the `ports/executorch` directory to your VCPKG installation's `ports/` directory
+3. Update the `SHA512` hash in `portfile.cmake` after the first build attempt
+4. Test the port locally: `vcpkg install executorch`
+5. Submit a pull request to the VCPKG repository
+
+### For Local Development
+
+To use this port locally without submitting to the registry:
+
+1. Copy the `ports/executorch` directory to your VCPKG installation's `ports/` directory
+2. Install: `vcpkg install executorch`
+3. Use in your project as shown in the `usage` file
+
+## Features
+
+The port supports several optional features:
+
+- **xnnpack**: XNNPACK backend for accelerated inference
+- **coreml**: CoreML backend (Apple platforms only)
+- **mps**: Metal Performance Shaders backend (Apple platforms only)
+- **vulkan**: Vulkan backend for GPU acceleration
+- **qnn**: Qualcomm QNN backend
+- **portable-ops**: Portable CPU operators
+- **optimized-ops**: Optimized CPU operators
+- **quantized-ops**: Quantized operators
+- **pybind**: Python bindings
+- **tests**: Build and run tests
+
+To install with features:
+```bash
+vcpkg install executorch[portable-ops,xnnpack]
+```
+
+## Maintenance
+
+When a new version of ExecuTorch is released:
+
+1. Update the `version-string` in `vcpkg.json`
+2. Update the `REF` in `portfile.cmake` to point to the new release tag
+3. Run `vcpkg install executorch` and update the `SHA512` hash based on the error message
+4. Test the build
+5. Update the VCPKG registry
+
+## Documentation
+
+- ExecuTorch: https://pytorch.org/executorch/
+- VCPKG Getting Started: https://learn.microsoft.com/en-us/vcpkg/get_started/get-started
+- VCPKG Packaging Guide: https://learn.microsoft.com/en-us/vcpkg/get_started/get-started-packaging
diff --git a/ports/executorch/portfile.cmake b/ports/executorch/portfile.cmake
new file mode 100644
index 00000000000..e3b61a74075
--- /dev/null
+++ b/ports/executorch/portfile.cmake
@@ -0,0 +1,84 @@
+vcpkg_check_linkage(ONLY_STATIC_LIBRARY)
+
+vcpkg_from_github(
+    OUT_SOURCE_PATH SOURCE_PATH
+    REPO pytorch/executorch
+    REF "v${VERSION}"
+    SHA512 0
+    HEAD_REF main
+)
+
+# Set CMake options based on features
+set(FEATURE_OPTIONS "")
+
+if("xnnpack" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_XNNPACK=ON)
+endif()
+
+if("coreml" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_COREML=ON)
+endif()
+
+if("mps" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_MPS=ON)
+endif()
+
+if("vulkan" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_VULKAN=ON)
+endif()
+
+if("qnn" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_QNN=ON)
+endif()
+
+if("portable-ops" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_PORTABLE_OPS=ON)
+endif()
+
+if("optimized-ops" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_OPTIMIZED_OPS=ON)
+endif()
+
+if("quantized-ops" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_QUANTIZED_OPS=ON)
+endif()
+
+if("pybind" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_PYBIND=ON)
+endif()
+
+if("tests" IN_LIST FEATURES)
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_TESTS=ON)
+else()
+    list(APPEND FEATURE_OPTIONS -DEXECUTORCH_BUILD_TESTS=OFF)
+endif()
+
+vcpkg_cmake_configure(
+    SOURCE_PATH "${SOURCE_PATH}"
+    OPTIONS
+        ${FEATURE_OPTIONS}
+        -DEXECUTORCH_ENABLE_LOGGING=ON
+        -DEXECUTORCH_ENABLE_PROGRAM_VERIFICATION=ON
+        -DEXECUTORCH_BUILD_EXTENSION_DATA_LOADER=ON
+        -DEXECUTORCH_BUILD_EXTENSION_MODULE=ON
+        -DEXECUTORCH_BUILD_EXTENSION_RUNNER_UTIL=ON
+)
+
+vcpkg_cmake_build()
+
+if("tests" IN_LIST FEATURES)
+    vcpkg_cmake_test()
+endif()
+
+vcpkg_cmake_install()
+
+vcpkg_cmake_config_fixup(CONFIG_PATH lib/cmake/ExecuTorch)
+
+file(REMOVE_RECURSE "${CURRENT_PACKAGES_DIR}/debug/include")
+file(REMOVE_RECURSE "${CURRENT_PACKAGES_DIR}/debug/share")
+
+# Handle copyright
+vcpkg_install_copyright(FILE_LIST "${SOURCE_PATH}/LICENSE")
+
+# Create usage file
+file(INSTALL "${CMAKE_CURRENT_LIST_DIR}/usage" DESTINATION "${CURRENT_PACKAGES_DIR}/share/${PORT}")
diff --git a/ports/executorch/usage b/ports/executorch/usage
new file mode 100644
index 00000000000..1759ffced67
--- /dev/null
+++ b/ports/executorch/usage
@@ -0,0 +1,30 @@
+ExecuTorch provides CMake targets:
+
+    find_package(ExecuTorch CONFIG REQUIRED)
+    target_link_libraries(main PRIVATE executorch)
+
+Available components/backends can be linked as needed:
+- executorch - Core runtime library
+- extension_module - Extension module support
+- extension_data_loader - Data loader extension
+- extension_runner_util - Runner utilities
+- portable_ops_lib - Portable CPU operators (if portable-ops feature enabled)
+- optimized_ops_lib - Optimized CPU operators (if optimized-ops feature enabled)
+- quantized_ops_lib - Quantized operators (if quantized-ops feature enabled)
+- xnnpack_backend - XNNPACK backend (if xnnpack feature enabled)
+- coreml_backend - CoreML backend (if coreml feature enabled, Apple only)
+- mps_backend - MPS backend (if mps feature enabled, Apple only)
+- vulkan_backend - Vulkan backend (if vulkan feature enabled)
+- qnn_backend - Qualcomm QNN backend (if qnn feature enabled)
+
+Example with specific components:
+
+    find_package(ExecuTorch CONFIG REQUIRED)
+    target_link_libraries(main PRIVATE 
+        executorch
+        extension_module
+        portable_ops_lib
+    )
+
+For more information, see:
+https://pytorch.org/executorch/
diff --git a/ports/executorch/vcpkg.json b/ports/executorch/vcpkg.json
new file mode 100644
index 00000000000..9415e668ddc
--- /dev/null
+++ b/ports/executorch/vcpkg.json
@@ -0,0 +1,64 @@
+{
+  "name": "executorch",
+  "version-string": "1.2.0-alpha",
+  "description": "ExecuTorch: On-device AI across mobile, embedded and edge for PyTorch. PyTorch's unified on-device AI inference framework for deploying models from smartphones to microcontrollers.",
+  "homepage": "https://pytorch.org/executorch/",
+  "license": "BSD-3-Clause",
+  "supports": "!uwp",
+  "dependencies": [
+    {
+      "name": "vcpkg-cmake",
+      "host": true
+    },
+    {
+      "name": "vcpkg-cmake-config",
+      "host": true
+    }
+  ],
+  "features": {
+    "xnnpack": {
+      "description": "Build with XNNPACK backend support",
+      "dependencies": []
+    },
+    "coreml": {
+      "description": "Build with CoreML backend support (Apple platforms only)",
+      "dependencies": []
+    },
+    "mps": {
+      "description": "Build with MPS (Metal Performance Shaders) backend support (Apple platforms only)",
+      "dependencies": []
+    },
+    "vulkan": {
+      "description": "Build with Vulkan backend support",
+      "dependencies": []
+    },
+    "qnn": {
+      "description": "Build with Qualcomm QNN backend support",
+      "dependencies": []
+    },
+    "portable-ops": {
+      "description": "Build with portable CPU operators",
+      "dependencies": []
+    },
+    "optimized-ops": {
+      "description": "Build with optimized CPU operators",
+      "dependencies": []
+    },
+    "quantized-ops": {
+      "description": "Build with quantized operators",
+      "dependencies": []
+    },
+    "pybind": {
+      "description": "Build Python bindings",
+      "dependencies": [
+        "pybind11"
+      ]
+    },
+    "tests": {
+      "description": "Build tests",
+      "dependencies": [
+        "gtest"
+      ]
+    }
+  }
+}