diff --git a/docs/how-to/model-performance-tuning.rst b/docs/how-to/model-performance-tuning.rst
new file mode 100644
index 00000000000..745abea8211
--- /dev/null
+++ b/docs/how-to/model-performance-tuning.rst
@@ -0,0 +1,129 @@
+.. meta::
+  :description: Tune MIGraphX model performance with environment variables
+  :keywords: MIGraphX, performance, tuning, environment variables, ROCm
+
+********************************************************************
+Tune model performance
+********************************************************************
+
+MIGraphX exposes environment variables that change compilation behavior and
+kernel selection at runtime. Set these variables before you compile or run a
+model to experiment with performance options without changing application
+code.
+
+For the complete list of environment variables, including developer-only
+tracing and debugging options, see
+:doc:`MIGraphX environment variables <../reference/MIGraphX-dev-env-vars>`.
+
+Set environment variables
+====================================================================
+
+Export a variable in your shell before running ``migraphx-driver`` or your
+application:
+
+.. code-block:: shell
+
+   export MIGRAPHX_SET_GEMM_PROVIDER=hipblaslt
+   /opt/rocm/bin/migraphx-driver perf model.onnx --onnx --gpu
+
+Replace ``model.onnx`` with your model path.
+
+Variables that take comma-separated lists must not include spaces between
+entries. Quote the value when your shell requires it.
+
+Common tuning variables
+====================================================================
+
+The following variables are the most commonly used for model performance
+tuning. Each entry describes accepted values and default behavior documented
+in the MIGraphX source tree.
+
+.. list-table::
+   :widths: 35 65
+   :header-rows: 1
+
+   * - Environment variable
+     - Values
+   * - ``MIGRAPHX_GPU_OPTIONS``
+     - JSON object of backend options (for example, ``{convolution_layout:channels_last}``). Quotes around keys and values are optional. Unrecognized options are ignored.
+   * - ``MIGRAPHX_SET_GEMM_PROVIDER``
+     - ``hipblaslt`` or ``rocblas``. Default: ``rocblas`` on gfx90a; ``hipblaslt`` on all other architectures.
+   * - ``MIGRAPHX_ENABLE_GEMM_TUNING``
+     - ``1`` enables exhaustive GEMM tuning even when ``--exhaustive-tune`` is not set. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_ENABLE_HIP_GEMM_TUNING``
+     - ``1`` enables exhaustive hipBLASLt tuning. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_MLIR_TUNE_EXHAUSTIVE``
+     - ``1`` enables exhaustive MLIR tuning. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_MLIR_USE_SPECIFIC_OPS``
+     - Comma-separated list of operations (for example, ``attention``, ``convolution``, ``dot``, ``fused_dot``, ``fused_convolution``, ``fused``). Prefix an entry with ``~`` to negate it.
+   * - ``MIGRAPHX_DISABLE_MLIR``
+     - ``1`` disables rocMLIR. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_ENABLE_CK``
+     - ``1`` enables Composable Kernel. Use with ``MIGRAPHX_DISABLE_MLIR=1``.
+   * - ``MIGRAPHX_USE_FAST_SOFTMAX``
+     - ``1`` enables fast softmax optimization. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_ENABLE_LAYERNORM_FUSION``
+     - ``1`` enables layernorm fusion. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_FLASH_DECODING_ENABLED``
+     - ``1`` enables flash decoding for attention fusion. Default: ``0``.
+   * - ``MIGRAPHX_SKIP_BENCHMARKING``
+     - ``1`` skips kernel benchmarking and compiles with the first available solution. ``0`` returns to default behavior.
+
+Tuning examples
+====================================================================
+
+Select a GEMM provider
+--------------------------------------------------------------------
+
+Set the general matrix multiply (GEMM) provider before compilation:
+
+.. code-block:: shell
+
+   export MIGRAPHX_SET_GEMM_PROVIDER=hipblaslt
+   /opt/rocm/bin/migraphx-driver perf model.onnx --onnx --gpu
+
+Replace ``model.onnx`` with your model path.
+
+Enable exhaustive GEMM tuning
+--------------------------------------------------------------------
+
+Search for the fastest GEMM kernel configuration:
+
+.. code-block:: shell
+
+   export MIGRAPHX_ENABLE_GEMM_TUNING=1
+   /opt/rocm/bin/migraphx-driver perf model.onnx --onnx --gpu --exhaustive-tune
+
+You can also pass ``--exhaustive-tune`` on the command line. The environment
+variable enables exhaustive GEMM tuning even when that flag is omitted.
+
+Specify MLIR operations
+--------------------------------------------------------------------
+
+Force specific operations to lower through MLIR. The list is comma-separated:
+
+.. code-block:: shell
+
+   export MIGRAPHX_MLIR_USE_SPECIFIC_OPS=attention,dot
+   /opt/rocm/bin/migraphx-driver compile model.onnx --onnx --gpu --text
+
+Replace ``model.onnx`` with your model path.
+
+Measure the effect
+====================================================================
+
+Compare performance before and after changing a variable:
+
+.. code-block:: shell
+
+   /opt/rocm/bin/migraphx-driver perf model.onnx --onnx --gpu -n 50
+
+Replace ``model.onnx`` with your model path and adjust ``-n`` for the number
+of timing iterations.
+
+See also
+====================================================================
+
+* :doc:`MIGraphX driver <../migraphx-driver>` for ``perf`` and ``compile`` commands.
+* :doc:`MIGraphX environment variables <../reference/MIGraphX-dev-env-vars>` for
+  the full variable reference, including pass controls and compilation tracing.
diff --git a/docs/how-to/model-validation.rst b/docs/how-to/model-validation.rst
new file mode 100644
index 00000000000..1e26b80f502
--- /dev/null
+++ b/docs/how-to/model-validation.rst
@@ -0,0 +1,159 @@
+.. meta::
+  :description: Validate MIGraphX model outputs
+  :keywords: MIGraphX, validation, verify, environment variables, ROCm
+
+********************************************************************
+Validate model outputs
+********************************************************************
+
+Use MIGraphX validation tools to compare GPU or CPU target output against the
+reference implementation. Validation helps confirm that compilation and
+quantization changes produce numerically consistent results.
+
+Validate with migraphx-driver
+====================================================================
+
+The ``verify`` command compiles your model for the reference and target
+backends, runs both, and checks that outputs match within configured
+tolerances. See :doc:`MIGraphX driver <../migraphx-driver>` for full command
+documentation.
+
+Basic verification
+--------------------------------------------------------------------
+
+Verify an ONNX model against the default target:
+
+.. code-block:: shell
+
+   /opt/rocm/bin/migraphx-driver verify model.onnx --onnx
+
+Replace ``model.onnx`` with your model path.
+
+Verify on GPU with tolerance settings
+--------------------------------------------------------------------
+
+Set absolute tolerance (``atol``), relative tolerance (``rtol``), and
+root-mean-square tolerance (``rms-tol``) when comparing outputs:
+
+.. code-block:: shell
+
+   /opt/rocm/bin/migraphx-driver verify model.onnx --onnx --gpu \
+       --atol 1e-5 --rtol 1e-5 --rms-tol 0.001
+
+Replace ``model.onnx`` with your model path. Default tolerance values are
+``0.001`` for ``atol``, ``rtol``, and ``rms-tol``.
+
+Verify after quantization
+--------------------------------------------------------------------
+
+Check that fp16 quantization preserves accuracy within your tolerances:
+
+.. code-block:: shell
+
+   /opt/rocm/bin/migraphx-driver verify model.onnx --onnx --gpu --fp16 \
+       --atol 1e-3 --rtol 1e-3
+
+Replace ``model.onnx`` with your model path.
+
+Additional verify options
+--------------------------------------------------------------------
+
+The driver supports these validation-related options:
+
+.. list-table::
+   :widths: 30 70
+   :header-rows: 1
+
+   * - Option
+     - Description
+   * - ``--per-instruction`` or ``-i``
+     - Verifies each instruction individually.
+   * - ``--reduce`` or ``-r``
+     - Reduces the program and verifies the reduced graph.
+   * - ``--atol``
+     - Sets tolerance for elementwise absolute difference (default: ``0.001``).
+   * - ``--rtol``
+     - Sets tolerance for elementwise relative difference (default: ``0.001``).
+   * - ``--rms-tol``
+     - Sets tolerance for root-mean-square error (default: ``0.001``).
+
+Validation environment variables
+====================================================================
+
+Set these environment variables before running ``migraphx-driver verify`` or
+your application to adjust validation behavior.
+
+Output comparison
+--------------------------------------------------------------------
+
+.. list-table::
+   :widths: 35 65
+   :header-rows: 1
+
+   * - Environment variable
+     - Values
+   * - ``MIGRAPHX_VERIFY_ENABLE_ALLCLOSE``
+     - ``1`` verifies range tolerance using ``allclose``. ``0`` returns to default behavior.
+   * - ``MIGRAPHX_VERIFY_DUMP_DIFF``
+     - ``1`` writes test output and reference output when they differ. ``0`` returns to default behavior.
+
+Example:
+
+.. code-block:: shell
+
+   export MIGRAPHX_VERIFY_DUMP_DIFF=1
+   /opt/rocm/bin/migraphx-driver verify model.onnx --onnx --gpu
+
+Replace ``model.onnx`` with your model path.
+
+Testing and debugging
+--------------------------------------------------------------------
+
+These variables apply when running MIGraphX verify tests or debugging
+validation failures during development:
+
+.. list-table::
+   :widths: 35 65
+   :header-rows: 1
+
+   * - Environment variable
+     - Values
+   * - ``MIGRAPHX_TRACE_TEST``
+     - ``1`` prints reference and target programs even when verify tests pass.
+   * - ``MIGRAPHX_TRACE_TEST_COMPILE``
+     - ``cpu`` or ``gpu`` turns on compile tracing for verify tests on the given target. Cannot be used with ``MIGRAPHX_TRACE_COMPILE``.
+   * - ``MIGRAPHX_DUMP_TEST``
+     - ``1`` writes the model under verification to an MXR file.
+
+Example:
+
+.. code-block:: shell
+
+   export MIGRAPHX_VERIFY_DUMP_DIFF=1
+   export MIGRAPHX_TRACE_TEST=1
+   /opt/rocm/bin/migraphx-driver verify model.onnx --onnx --gpu
+
+Replace ``model.onnx`` with your model path.
+
+Graph validation during development
+--------------------------------------------------------------------
+
+When developing passes or matchers, enable module validation after pattern
+matches:
+
+.. code-block:: shell
+
+   export MIGRAPHX_VALIDATE_MATCHES=1
+
+See :doc:`MIGraphX environment variables <../reference/MIGraphX-dev-env-vars>`
+for the complete list of validation, testing, and tracing variables.
+
+See also
+====================================================================
+
+* :doc:`MIGraphX driver <../migraphx-driver>` for the ``verify`` command and
+  tolerance options.
+* :doc:`Precision support <../reference/MIGraphX-data-type-support>` for
+  supported data types when validating quantized models.
+* :doc:`Tune model performance <./model-performance-tuning>` when adjusting
+  compilation options that can affect numerical output.
diff --git a/docs/index.rst b/docs/index.rst
index d26846ffc58..8ff70d07817 100644
--- a/docs/index.rst
+++ b/docs/index.rst
@@ -13,17 +13,28 @@ It compiles trained models from end-to-end to optimize for inference performance
 
 The MIGraphX public repository is located at `https://github.com/ROCm/AMDMIGraphX/ <https://github.com/ROCm/AMDMIGraphX/>`__.
 
-You can integrate MIGraphX with PyTorch worflows by using the Torch-MIGraphX library.
+You can integrate MIGraphX with PyTorch workflows by using the Torch-MIGraphX library.
 The public repository is located at `https://github.com/ROCm/torch_migraphx/ <https://github.com/ROCm/torch_migraphx/>`__.
 
 .. grid:: 2
   :gutter: 3
 
+  .. grid-item-card:: Getting started
+
+    * :doc:`Get started with MIGraphX <./tutorials/getting-started>`
+    * :doc:`Parse, load, and save a model <./tutorials/parse-load-save-tutorial>`
+
   .. grid-item-card:: Install
 
     * :doc:`MIGraphX on ROCm installation <./install/install-migraphx>`
+    * :doc:`Install MIGraphX with Docker <./install/install-docker>`
     * :doc:`Torch-MIGraphX installation <./install/install-torch-migraphx>`
 
+  .. grid-item-card:: How-to
+
+    * :doc:`Tune model performance <./how-to/model-performance-tuning>`
+    * :doc:`Validate model outputs <./how-to/model-validation>`
+
   .. grid-item-card:: Conceptual
 
     * :doc:`Deep learning compilation with MIGraphX <./conceptual/deep-learning-compilation>`
diff --git a/docs/install/install-docker.rst b/docs/install/install-docker.rst
new file mode 100644
index 00000000000..45f7dd41c1d
--- /dev/null
+++ b/docs/install/install-docker.rst
@@ -0,0 +1,106 @@
+.. meta::
+  :description: Install MIGraphX using Docker
+  :keywords: install, Docker, MIGraphX, AMD, ROCm, container
+
+********************************************************************
+Install MIGraphX with Docker
+********************************************************************
+
+Docker provides a development environment with MIGraphX build prerequisites
+preinstalled. Use this approach when you want to build MIGraphX from source
+without installing dependencies on your host system.
+
+The default ``Dockerfile`` at the repository root builds against ROCm 7.13 and
+newer using TheRock (``amdrocm-*``) packages. Alternative Dockerfiles for
+other ROCm releases are available under ``tools/docker/``.
+
+Prerequisites
+====================================================================
+
+Before you build or run a MIGraphX Docker image, install the following on
+your host:
+
+* `ROCm <https://rocm.docs.amd.com/projects/install-on-linux/en/latest/>`__
+  with GPU drivers configured
+* Docker with access to ``/dev/kfd`` and ``/dev/dri``
+
+Build the default Docker image
+====================================================================
+
+From the repository root, build the image:
+
+.. code-block:: shell
+
+   docker build -t migraphx .
+
+To reduce image size, set ``GPU_ARCH`` to the GPU architecture on your
+system. Leaving ``GPU_ARCH`` unset installs device code for all supported
+architectures, which is useful when the same image runs on different
+ROCm-supported GPUs.
+
+.. code-block:: shell
+
+   docker build -t migraphx --build-arg GPU_ARCH=$(rocminfo | grep -o -m1 'gfx.*') .
+
+The root ``Dockerfile`` accepts the following build arguments:
+
+.. list-table::
+   :widths: 30 70
+   :header-rows: 1
+
+   * - Build argument
+     - Description
+   * - ``ROCM_VERSION``
+     - ROCm release version for versioned package names (default: ``7.14``).
+   * - ``GPU_ARCH``
+     - GPU architecture family (for example, ``gfx942``, ``gfx120x``). Leave empty for arch-independent packages.
+   * - ``USE_WHL``
+     - When set to a non-empty value, installs ROCm from Python wheels instead of system packages.
+   * - ``INDEX_URL``
+     - pip index URL used when ``USE_WHL`` is set (default: ``https://repo.amd.com/rocm/whl-multi-arch/``).
+
+Run the container
+====================================================================
+
+Start an interactive container with your repository mounted at
+``/code/AMDMIGraphX``:
+
+.. code-block:: shell
+
+   docker run --device='/dev/kfd' --device='/dev/dri' \
+       -v=`pwd`:/code/AMDMIGraphX -w /code/AMDMIGraphX \
+       --group-add video -it migraphx
+
+Build MIGraphX inside the container
+====================================================================
+
+Inside the container, prerequisites are already installed. Follow the steps in
+:doc:`MIGraphX on ROCm installation <./install-migraphx>` to build from source,
+starting from the rbuild or CMake build steps.
+
+Alternative Dockerfiles
+====================================================================
+
+Use an alternative Dockerfile when your target ROCm release requires a
+different base image or package source.
+
+.. list-table::
+   :widths: 35 65
+   :header-rows: 1
+
+   * - Dockerfile path
+     - Use when
+   * - ``tools/docker/legacy.dockerfile``
+     - Building against ROCm 7.2.x and older on Ubuntu 22.04.
+   * - ``tools/docker/ubuntu_2404.dockerfile``
+     - Building on Ubuntu 24.04 with ROCm 7.1.1.
+   * - ``tools/docker/ubuntu_2204.dockerfile``
+     - Building on Ubuntu 22.04 with ROCm 6.4.2.
+
+Build with the ``-f`` flag to select a Dockerfile. For example:
+
+.. code-block:: shell
+
+   docker build -t migraphx:legacy -f tools/docker/legacy.dockerfile .
+
+Then follow the same ``docker run`` and build steps described above.
diff --git a/docs/install/install-migraphx.rst b/docs/install/install-migraphx.rst
index 3eaaa4ed64b..f206c8e0542 100644
--- a/docs/install/install-migraphx.rst
+++ b/docs/install/install-migraphx.rst
@@ -12,7 +12,10 @@ for instructions.
 
 Installing MIGraphX using a package installer is sufficient for most users 
 who want to use the MIGraphX API. If you plan to develop for MIGraphX or 
-contribute to the source code, see `Developing for MIGraphX <../dev/contributing-to-migraphx.html>`_
+contribute to the source code, see `Developing for MIGraphX <../dev/contributing-to-migraphx.html>`_.
+
+For a containerized build environment with prerequisites preinstalled, see
+:doc:`Install MIGraphX with Docker <./install-docker>`.
 
 Install MIGraphX with a package installer
 ====================================================================
diff --git a/docs/reference/MIGraphX-dev-env-vars.rst b/docs/reference/MIGraphX-dev-env-vars.rst
index 746f5b62618..b5e4e3e9b45 100644
--- a/docs/reference/MIGraphX-dev-env-vars.rst
+++ b/docs/reference/MIGraphX-dev-env-vars.rst
@@ -8,6 +8,10 @@ MIGraphX environment variables
 
 The MIGraphX environment variables can be used by contributors to the MIGraphX code base to customize tuning, verification, and tracing.
 
+For user-facing guides on the most common tuning and validation variables, see
+:doc:`Tune model performance <../how-to/model-performance-tuning>` and
+:doc:`Validate model outputs <../how-to/model-validation>`.
+
 
 Model performance tunable variables
 ************************************
diff --git a/docs/reference/MIGraphX-user-reference.rst b/docs/reference/MIGraphX-user-reference.rst
index f88208bdc0e..69e3f178ff7 100644
--- a/docs/reference/MIGraphX-user-reference.rst
+++ b/docs/reference/MIGraphX-user-reference.rst
@@ -6,6 +6,12 @@
 MIGraphX user reference
 ========================================================
 
-The MIGraphX library includes :doc:`C++ <./MIGraphX-cpp>` and :doc:`Python <./MIGraphX-py>` APIs for creating applications. 
+The MIGraphX library includes :doc:`C++ <./MIGraphX-cpp>` and :doc:`Python <./MIGraphX-py>` APIs for creating applications.
+
+Operational guides:
+
+* :doc:`Tune model performance <../how-to/model-performance-tuning>`
+* :doc:`Validate model outputs <../how-to/model-validation>`
+* :doc:`MIGraphX driver <../migraphx-driver>`
 
 To contribute to the MIGraphX code base, see :doc:`the MIGraphX contributor documentation <./MIGraphX-dev-reference>`.
\ No newline at end of file
diff --git a/docs/sphinx/_toc.yml.in b/docs/sphinx/_toc.yml.in
index 79e4210f8a6..78b4912e350 100644
--- a/docs/sphinx/_toc.yml.in
+++ b/docs/sphinx/_toc.yml.in
@@ -3,9 +3,18 @@
 root: index
 subtrees:
 
+  - caption: Getting started
+    entries:
+    - file: tutorials/getting-started
+      title: Get started with MIGraphX
+    - file: tutorials/parse-load-save-tutorial
+      title: Parse, load, and save a model
+
   - caption: Installation
     entries: 
     - file: install/install-migraphx
+    - file: install/install-docker
+      title: Install MIGraphX with Docker
     - file: install/install-torch-migraphx
 
   - caption: Conceptual
@@ -13,6 +22,13 @@ subtrees:
     - file: conceptual/deep-learning-compilation
       title: Deep learning compilation with MIGraphX
     
+  - caption: How-to
+    entries:
+    - file: how-to/model-performance-tuning
+      title: Tune model performance
+    - file: how-to/model-validation
+      title: Validate model outputs
+
   - caption: Reference
     entries:
     - file: reference/MIGraphX-user-reference
diff --git a/docs/tutorials/MIGraphX-examples.rst b/docs/tutorials/MIGraphX-examples.rst
index f93f86ce8e4..99570f87b83 100644
--- a/docs/tutorials/MIGraphX-examples.rst
+++ b/docs/tutorials/MIGraphX-examples.rst
@@ -1,16 +1,204 @@
 .. meta::
-  :description: MIGraphX Examples
+  :description: MIGraphX examples
   :keywords: MIGraphX, AMD, ROCm, examples
 
 ********************************************************************
 MIGraphX examples
 ********************************************************************
 
-Example code for the following use cases is available in the `MIGraphX GitHub repository <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples>`_:
+The MIGraphX repository includes examples for common inference workflows.
+Each example below is documented in its README file. Follow the linked path
+in the `MIGraphX GitHub repository <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples>`__
+to build and run the example.
 
-* `Diffusion inference <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/diffusion>`_
-* `AMD MIGraphX usage and utilities <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx>`_
-* `Natural language processing inference <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/nlp>`_
-* `ONNX runtime <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/onnxruntime>`_
-* `Transformer inference <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/transformers>`_
-* `Vision inference <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision>`_
\ No newline at end of file
+To swap in your own model, replace the model file path named in the example
+README or source code with your ONNX or serialized MIGraphX program. Adjust
+input dimensions and data loading to match your model's parameter shapes.
+
+Getting started examples
+====================================================================
+
+These examples teach core MIGraphX workflows. Start here if you are new to
+the library.
+
+.. list-table::
+   :widths: 25 35 40
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - :doc:`Parse, load, and save a model <./parse-load-save-tutorial>`
+     - C++ application developers
+     - Parse ONNX models, inspect graphs, and save or load programs in MessagePack or JSON format.
+   * - `cpp_mnist <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/cpp_mnist>`__
+     - C++ application developers
+     - End-to-end inference: parse ONNX, compile for GPU or CPU, quantize, prepare inputs, and evaluate.
+   * - `migraphx_driver <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/migraphx_driver>`__
+     - Model integrators
+     - Command-line usage of ``migraphx-driver`` for read, compile, run, verify, and perf without writing code.
+
+MIGraphX API examples
+====================================================================
+
+Examples under ``examples/migraphx/`` cover MIGraphX-specific utilities and
+advanced C++ API usage.
+
+.. list-table::
+   :widths: 30 35 35
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - `cpp_parse_load_save <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/cpp_parse_load_save>`__
+     - C++ application developers
+     - Parse, load, and save graph programs. Swap in your ONNX file as the input argument.
+   * - `cpp_dynamic_batch <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/cpp_dynamic_batch>`__
+     - C++ application developers
+     - Run programs with dynamic batch sizes using ``dynamic_dimension`` objects. Replace the ONNX input file with your dynamic-batch model.
+   * - `cpp_trace_callback <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/cpp_trace_callback>`__
+     - MIGraphX contributors and debuggers
+     - Inspect operator output buffers during evaluation with ``program::run_trace``.
+   * - `export_frozen_graph_tf1 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/export_frozen_graph_tf1>`__
+     - Model integrators
+     - Export frozen TensorFlow 1 graphs for MIGraphX ingestion.
+   * - `export_frozen_graph_tf2 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/export_frozen_graph_tf2>`__
+     - Model integrators
+     - Export frozen TensorFlow 2 graphs for MIGraphX ingestion.
+
+Vision examples
+====================================================================
+
+Examples under ``examples/vision/`` cover image classification, detection,
+segmentation, and super resolution.
+
+.. list-table::
+   :widths: 30 35 35
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - `cpp_mnist <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/cpp_mnist>`__
+     - C++ application developers
+     - MNIST digit classification with the C++ API.
+   * - `python_resnet50 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_resnet50>`__
+     - Python application developers
+     - ResNet50 V2 inference in Python via a Jupyter notebook. Replace the bundled model with your own pre-trained classification model.
+   * - `python_nfnet <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_nfnet>`__
+     - Python application developers
+     - NFNet inference and ONNX Runtime comparison.
+   * - `python_super_resolution <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_super_resolution>`__
+     - Python application developers
+     - Super resolution inference.
+   * - `python_unet <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_unet>`__
+     - Python application developers
+     - U-Net segmentation inference.
+   * - `python_3dunet <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_3dunet>`__
+     - Python application developers
+     - 3D U-Net inference.
+   * - `python_yolov4 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_yolov4>`__
+     - Python application developers
+     - YOLOv4 object detection inference.
+   * - `python_yolo26 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/python_yolo26>`__
+     - Python application developers
+     - YOLO26 object detection inference.
+
+Natural language processing examples
+====================================================================
+
+Examples under ``examples/nlp/`` cover question answering and speech
+recognition.
+
+.. list-table::
+   :widths: 30 35 35
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - `python_bert_squad <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/nlp/python_bert_squad>`__
+     - Python application developers
+     - BERT question answering on SQuAD. Replace ``bertsquad-10.onnx`` and ``inputs.json`` with your model and input data.
+   * - `python_rnnt <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/nlp/python_rnnt>`__
+     - Python application developers
+     - RNN-T speech recognition inference.
+
+Diffusion examples
+====================================================================
+
+Examples under ``examples/diffusion/`` cover text-to-image diffusion models.
+
+.. list-table::
+   :widths: 30 35 35
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - `python_stable_diffusion_21 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/diffusion/python_stable_diffusion_21>`__
+     - Python application developers
+     - Stable Diffusion 2.1 text-to-image inference.
+   * - `python_stable_diffusion_xl <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/diffusion/python_stable_diffusion_xl>`__
+     - Python application developers
+     - Stable Diffusion XL text-to-image inference.
+   * - `python_stable_diffusion_3 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/diffusion/python_stable_diffusion_3>`__
+     - Python application developers
+     - Stable Diffusion 3 text-to-image inference.
+   * - `python_flux <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/diffusion/python_flux>`__
+     - Python application developers
+     - Flux text-to-image inference.
+   * - `python_controlnet_canny_sd_15 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/diffusion/python_controlnet_canny_sd_15>`__
+     - Python application developers
+     - ControlNet Canny edge guidance with Stable Diffusion 1.5.
+
+Transformer examples
+====================================================================
+
+Examples under ``examples/transformers/`` cover large language model and
+speech model inference.
+
+.. list-table::
+   :widths: 30 35 35
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - `python_llama2 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/transformers/python_llama2>`__
+     - Python application developers
+     - Llama 2 inference.
+   * - `python_whisper <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/transformers/python_whisper>`__
+     - Python application developers
+     - Whisper automatic speech recognition.
+
+ONNX Runtime examples
+====================================================================
+
+Examples under ``examples/onnxruntime/`` show MIGraphX integration through
+the ONNX Runtime execution provider.
+
+.. list-table::
+   :widths: 30 35 35
+   :header-rows: 1
+
+   * - Example
+     - Target user
+     - What it demonstrates
+   * - `resnet50 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/onnxruntime/resnet50>`__
+     - ONNX Runtime users
+     - ResNet50 inference through ONNX Runtime with the MIGraphX execution provider.
+   * - `inceptionV3 <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/onnxruntime/inceptionV3>`__
+     - ONNX Runtime users
+     - InceptionV3 inference through ONNX Runtime with the MIGraphX execution provider.
+
+See also
+====================================================================
+
+* :doc:`Get started with MIGraphX <./getting-started>` for persona-based
+  starting paths.
+* :doc:`Install MIGraphX with Docker <../install/install-docker>` for a
+  containerized build environment.
+* :doc:`Validate model outputs <../how-to/model-validation>` to check example
+  output against the reference implementation.
diff --git a/docs/tutorials/getting-started.rst b/docs/tutorials/getting-started.rst
new file mode 100644
index 00000000000..e4a242e6c41
--- /dev/null
+++ b/docs/tutorials/getting-started.rst
@@ -0,0 +1,132 @@
+.. meta::
+  :description: Get started with MIGraphX after installation
+  :keywords: MIGraphX, getting started, tutorial, ROCm
+
+********************************************************************
+Get started with MIGraphX
+********************************************************************
+
+After you install MIGraphX, use this guide to choose a starting path based on
+how you plan to use the library. Each section links to validated examples and
+reference material in this repository.
+
+Choose your path
+====================================================================
+
+MIGraphX supports several integration patterns. Pick the section that matches
+your role and goal.
+
+Application developer using the C++ API
+====================================================================
+
+Use this path when you embed MIGraphX directly in a C++ application.
+
+1. Confirm MIGraphX is installed. See :doc:`MIGraphX on ROCm installation <../install/install-migraphx>`.
+2. Set ``CMAKE_PREFIX_PATH`` to your MIGraphX installation location and link
+   against the C++ API:
+
+   .. code-block:: cmake
+
+      find_package(migraphx)
+      target_link_libraries(my_app migraphx::c)
+
+   Replace ``my_app`` with your CMake target name.
+3. Work through :doc:`Parse, load, and save a model <./parse-load-save-tutorial>`
+   to learn how to load ONNX models and serialize programs.
+4. Continue with the C++ MNIST inference example at
+   ``examples/vision/cpp_mnist/`` in the
+   `MIGraphX GitHub repository <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/vision/cpp_mnist>`__
+   for a complete parse, compile, and evaluate workflow.
+5. Consult the :ref:`cpp-api-reference` for API details.
+
+Application developer using the Python API
+====================================================================
+
+Use this path when you call MIGraphX from Python.
+
+1. Confirm MIGraphX is installed. See :doc:`MIGraphX on ROCm installation <../install/install-migraphx>`.
+2. Add the Python module to your environment:
+
+   .. code-block:: shell
+
+      export PYTHONPATH=/opt/rocm/lib:$PYTHONPATH
+
+   Replace ``/opt/rocm/lib`` with the directory that contains the MIGraphX
+   Python module if your installation path differs.
+3. Browse the Python examples under ``examples/`` in the
+   `MIGraphX GitHub repository <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples>`__.
+   Start with a domain that matches your model, such as vision or natural
+   language processing.
+4. Consult the :ref:`python-api-reference` for API details.
+
+Model integrator or inference engineer
+====================================================================
+
+Use this path when you want to inspect, compile, verify, or benchmark models
+without writing application code.
+
+1. Confirm ``migraphx-driver`` is available at ``/opt/rocm/bin/migraphx-driver``,
+   or at ``AMDMIGraphX/build/bin/migraphx-driver`` after a source build.
+2. Read :doc:`MIGraphX driver <../migraphx-driver>` for commands such as
+   ``read``, ``compile``, ``run``, ``verify``, and ``perf``.
+3. Validate a model against the reference implementation:
+
+   .. code-block:: shell
+
+      /opt/rocm/bin/migraphx-driver verify model.onnx --onnx
+
+4. Measure performance:
+
+   .. code-block:: shell
+
+      /opt/rocm/bin/migraphx-driver perf model.onnx --onnx --gpu -n 50
+
+5. See :doc:`Validate model outputs <../how-to/model-validation>` and
+   :doc:`Tune model performance <../how-to/model-performance-tuning>` for
+   environment variables and tolerance settings.
+
+PyTorch user
+====================================================================
+
+Use this path when you integrate MIGraphX with PyTorch workflows through
+Torch-MIGraphX.
+
+1. Install Torch-MIGraphX. See
+   :doc:`Torch-MIGraphX installation <../install/install-torch-migraphx>`.
+2. Refer to the Torch-MIGraphX repository at
+   `https://github.com/ROCm/torch_migraphx/ <https://github.com/ROCm/torch_migraphx/>`__
+   for model conversion and inference examples.
+
+ONNX Runtime user
+====================================================================
+
+Use this path when you run inference through the ONNX Runtime MIGraphX
+execution provider.
+
+1. Install MIGraphX. See :doc:`MIGraphX on ROCm installation <../install/install-migraphx>`.
+2. Follow the ResNet50 example at
+   ``examples/onnxruntime/resnet50/`` in the
+   `MIGraphX GitHub repository <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/onnxruntime/resnet50>`__
+   for setup and inference steps with ONNX Runtime.
+
+MIGraphX contributor
+====================================================================
+
+Use this path when you modify the MIGraphX source code or add operators and
+passes.
+
+1. Set up a build environment. See :doc:`Install MIGraphX with Docker <../install/install-docker>`
+   or :doc:`MIGraphX on ROCm installation <../install/install-migraphx>`.
+2. Read :doc:`Develop for the MIGraphX code base <../dev/contributing-to-migraphx>`.
+3. Consult :doc:`MIGraphX environment variables <../reference/MIGraphX-dev-env-vars>`
+   for tuning, tracing, and testing options used during development.
+
+Next steps
+====================================================================
+
+* :doc:`Parse, load, and save a model <./parse-load-save-tutorial>` walks
+  through your first C++ workflow with an ONNX model.
+* :doc:`MIGraphX examples <./MIGraphX-examples>` lists documented examples by
+  domain and target user.
+* :doc:`Deep learning compilation with MIGraphX <../conceptual/deep-learning-compilation>`
+  explains how MIGraphX optimizes inference graphs.
diff --git a/docs/tutorials/parse-load-save-tutorial.rst b/docs/tutorials/parse-load-save-tutorial.rst
new file mode 100644
index 00000000000..b68e9389e0f
--- /dev/null
+++ b/docs/tutorials/parse-load-save-tutorial.rst
@@ -0,0 +1,131 @@
+.. meta::
+  :description: Parse, load, and save MIGraphX programs
+  :keywords: MIGraphX, tutorial, ONNX, parse, load, save, C++
+
+********************************************************************
+Parse, load, and save a model
+********************************************************************
+
+This tutorial walks through the
+``examples/migraphx/cpp_parse_load_save`` example in the MIGraphX repository.
+You learn how to parse an ONNX model into a MIGraphX program, inspect the
+graph, save it to MessagePack format, and reload it later.
+
+The example source is at
+`examples/migraphx/cpp_parse_load_save <https://github.com/ROCm/AMDMIGraphX/tree/develop/examples/migraphx/cpp_parse_load_save>`__
+in the MIGraphX GitHub repository.
+
+What you will learn
+====================================================================
+
+* Parse an ONNX file into a ``migraphx::program``.
+* Save a program to MessagePack (``.mxr``) format.
+* Load a previously saved program from MessagePack or JSON format.
+* Swap in your own ONNX model by changing the input file path.
+
+Prerequisites
+====================================================================
+
+* MIGraphX installed. See :doc:`MIGraphX on ROCm installation <../install/install-migraphx>`.
+* A C++ compiler and CMake 3.5 or later.
+* An ONNX model file. The tutorial uses any compatible ONNX file you provide
+  as ``input_model.onnx``.
+
+Build the example
+====================================================================
+
+From the ``examples/migraphx/cpp_parse_load_save`` directory:
+
+.. code-block:: shell
+
+   mkdir build
+   cd build
+   cmake ..
+   make
+
+This produces an executable named ``parse_load_save``.
+
+Parse an ONNX model
+====================================================================
+
+By default, the program parses the input file as ONNX and prints the internal
+graph structure. Replace ``input_model.onnx`` with the path to your model:
+
+.. code-block:: shell
+
+   ./parse_load_save input_model.onnx --parse onnx
+
+The C++ API call used in the example is:
+
+.. code-block:: cpp
+
+   migraphx::program p;
+   migraphx::onnx_options options;
+   p = parse_onnx(input_file, options);
+
+Computation graphs saved in a compatible serialized format, such as ONNX, can
+be read by MIGraphX to create a runnable program.
+
+Save a program
+====================================================================
+
+Save the parsed program to MessagePack format. Replace ``output_name`` with
+your preferred base filename (the ``.mxr`` extension is added automatically):
+
+.. code-block:: shell
+
+   ./parse_load_save input_model.onnx --parse onnx --save output_name
+
+The program writes ``output_name.mxr`` in the current directory.
+
+Load a saved program
+====================================================================
+
+Load a program that was previously saved in MessagePack format:
+
+.. code-block:: shell
+
+   ./parse_load_save output_name.mxr --load msgpack
+
+To load a program saved in JSON format:
+
+.. code-block:: shell
+
+   ./parse_load_save saved_program.json --load json
+
+Use your own model and data
+====================================================================
+
+To adapt this example for your workflow:
+
+1. Replace ``input_model.onnx`` with the path to your ONNX model file.
+2. If your model has dynamic or unspecified dimensions, configure
+   ``migraphx::onnx_options`` before parsing. For example, set a default batch
+   size with ``options.set_default_dim_value(batch)``, where ``batch`` is your
+   batch size integer.
+3. After parsing, compile the program for your target and run inference. See
+   the C++ MNIST example at
+   ``examples/vision/cpp_mnist/`` for compile and evaluate steps.
+
+Command-line reference
+====================================================================
+
+The ``parse_load_save`` executable supports the following usage:
+
+.. code-block:: text
+
+   ./parse_load_save <input_file> [options]
+   options:
+           --parse onnx
+           --load  json/msgpack
+           --save  <output_file>
+
+Next steps
+====================================================================
+
+* Complete an end-to-end inference workflow in the C++ MNIST example
+  documented at ``examples/vision/cpp_mnist/``.
+* Use :doc:`MIGraphX driver <../migraphx-driver>` to inspect and verify your
+  model from the command line without writing C++ code.
+* See :doc:`MIGraphX examples <./MIGraphX-examples>` for additional examples
+  by domain.
diff --git a/examples/migraphx/migraphx_docker/README.md b/examples/migraphx/migraphx_docker/README.md
index 55571a42b88..3a1e4830e71 100644
--- a/examples/migraphx/migraphx_docker/README.md
+++ b/examples/migraphx/migraphx_docker/README.md
@@ -1,3 +1,6 @@
 # MIGraphX Dockerfile
 
-Instructions for building and running the MIGraphX docker container can be found [here](https://github.com/ROCmSoftwarePlatform/AMDMIGraphX/blob/develop/README.md#using-docker) in this project's top level README.
+Instructions for building and running the MIGraphX Docker container are in the
+[MIGraphX documentation](https://rocm.docs.amd.com/projects/AMDMIGraphX/en/latest/install/install-docker.html)
+and in the [Use Docker](https://github.com/ROCm/AMDMIGraphX/blob/develop/README.md#use-docker)
+section of the repository README.