From 08e302257d47b7b86afd0bd97fbdacce35796632 Mon Sep 17 00:00:00 2001
From: Yingge He <157551214+yinggeh@users.noreply.github.com>
Date: Thu, 12 Feb 2026 11:50:02 -0800
Subject: [PATCH] docs: Fix broken links in "Perf benchmarking and tuning"
 section (#8649)

---
 deploy/mlflow-triton-plugin/README.md         |  42 +-
 docs/client_guide/python.rst                  |   5 +-
 docs/client_guide/python_readme.rst           | 300 --------
 docs/contents.rst                             |   6 +-
 ...nt_by_backend.rst => quick_deployment.rst} |   5 +-
 docs/getting_started/quick_start.rst          | 175 -----
 docs/llm_features/speculative_decoding.rst    |  36 +-
 .../speculative_decoding_by_backend_type.rst  |  39 -
 docs/perf_benchmark/genai-perf-README.rst     | 686 ------------------
 docs/perf_benchmark/genai_perf.rst            |   5 +-
 docs/perf_benchmark/model-analyzer-README.rst | 203 ------
 docs/perf_benchmark/model_analyzer.rst        |  28 +-
 docs/perf_benchmark/perf-analyzer-README.rst  | 175 -----
 docs/perf_benchmark/perf_analyzer.rst         |  18 +-
 14 files changed, 67 insertions(+), 1656 deletions(-)
 delete mode 100644 docs/client_guide/python_readme.rst
 rename docs/getting_started/{quick_deployment_by_backend.rst => quick_deployment.rst} (95%)
 delete mode 100644 docs/getting_started/quick_start.rst
 delete mode 100644 docs/llm_features/speculative_decoding_by_backend_type.rst
 delete mode 100644 docs/perf_benchmark/genai-perf-README.rst
 delete mode 100644 docs/perf_benchmark/model-analyzer-README.rst
 delete mode 100644 docs/perf_benchmark/perf-analyzer-README.rst

diff --git a/deploy/mlflow-triton-plugin/README.md b/deploy/mlflow-triton-plugin/README.md
index c011194299..2fcfc035d5 100644
--- a/deploy/mlflow-triton-plugin/README.md
+++ b/deploy/mlflow-triton-plugin/README.md
@@ -1,5 +1,5 @@
 <!--
-# Copyright 2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# Copyright 2021-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions
@@ -50,7 +50,7 @@ to be deployed.
 
 The plugin can be installed from source using the following commands
 
-```
+```bash
 python setup.py install
 ```
 
@@ -89,7 +89,7 @@ as additional artifact which Triton will be used to serve the model. Otherwise,
 the server should be run with auto-complete feature enabled
 (`--strict-model-config=false`) to generate the model configuration.
 
-```
+```bash
 import mlflow.onnx
 import onnx
 model = onnx.load("examples/onnx_float32_int32_int32/1/model.onnx")
@@ -105,7 +105,7 @@ containing the model files following the
 [model layout](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_repository.md#repository-layout).
 Below is an example usage:
 
-```
+```bash
 cd /scripts
 
 python publish_model_to_mlflow.py --model_name onnx_float32_int32_int32 --model_directory <path-to-the-examples-directory>/onnx_float32_int32_int32 --flavor triton
@@ -117,7 +117,7 @@ Once a model is published and tracked in MLflow, it can be deployed to Triton
 via MLflow's deployments command, the following command will download the model
 to Triton's model repository and request Triton to load the model.
 
-```
+```bash
 mlflow deployments create -t triton --flavor triton --name onnx_float32_int32_int32 -m models:/onnx_float32_int32_int32/1
 ```
 
@@ -126,7 +126,7 @@ mlflow deployments create -t triton --flavor triton --name onnx_float32_int32_in
 After the model is deployed, the following command is the CLI usage to send
 inference request to a deployment.
 
-```
+```bash
 mlflow deployments predict -t triton --name onnx_float32_int32_int32 --input-path <path-to-the-examples-directory>/input.json --output-path output.json
 ```
 
@@ -145,15 +145,15 @@ MLflow deployments create API deploys a model to the Triton target, which will
 download the model to Triton's model repository and request Triton to load the
 model.
 
-To create a MLflow deployment using CLI:
+To create a MLflow deployment using CLI
 
-```
+```bash
 mlflow deployments create -t triton --flavor triton --name model_name -m models:/model_name/1
 ```
 
-To create a MLflow deployment using Python API:
+To create a MLflow deployment using Python API
 
-```
+```bash
 from mlflow.deployments import get_deploy_client
 client = get_deploy_client('triton')
 client.create_deployment("model_name", "models:/model_name/1", flavor="triton")
@@ -167,13 +167,13 @@ Triton to unload the model.
 
 To delete a MLflow deployment using CLI
 
-```
+```bash
 mlflow deployments delete -t triton --name model_name
 ```
 
-To delete a MLflow deployment using CLI
+To delete a MLflow deployment using Python API
 
-```
+```bash
 from mlflow.deployments import get_deploy_client
 client = get_deploy_client('triton')
 client.delete_deployment("model_name")
@@ -187,13 +187,13 @@ repository and request Triton to reload the model.
 
 To update a MLflow deployment using CLI
 
-```
+```bash
 mlflow deployments update -t triton --flavor triton --name model_name -m models:/model_name/2
 ```
 
 To update a MLflow deployment using Python API
 
-```
+```bash
 from mlflow.deployments import get_deploy_client
 client = get_deploy_client('triton')
 client.update_deployment("model_name", "models:/model_name/2", flavor="triton")
@@ -205,13 +205,13 @@ MLflow deployments list API lists all existing deployments in Triton target.
 
 To list all MLflow deployments using CLI
 
-```
+```bash
 mlflow deployments list -t triton
 ```
 
 To list all MLflow deployments using Python API
 
-```
+```bash
 from mlflow.deployments import get_deploy_client
 client = get_deploy_client('triton')
 client.list_deployments()
@@ -223,12 +223,12 @@ MLflow deployments get API returns information regarding a specific deployments
 in Triton target.
 
 To list a specific MLflow deployment using CLI
-```
+```bash
 mlflow deployments get -t triton --name model_name
 ```
 
 To list a specific MLflow deployment using Python API
-```
+```bash
 from mlflow.deployments import get_deploy_client
 client = get_deploy_client('triton')
 client.get_deployment("model_name")
@@ -241,14 +241,14 @@ request to Triton and returns the Triton response.
 
 To run inference using CLI
 
-```
+```bash
 mlflow deployments predict -t triton --name model_name --input-path input_file --output-path output_file
 
 ```
 
 To run inference using Python API
 
-```
+```bash
 from mlflow.deployments import get_deploy_client
 client = get_deploy_client('triton')
 client.predict("model_name", inputs)
diff --git a/docs/client_guide/python.rst b/docs/client_guide/python.rst
index 545f4f6042..e546f7b18e 100644
--- a/docs/client_guide/python.rst
+++ b/docs/client_guide/python.rst
@@ -1,5 +1,5 @@
 ..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+.. Copyright 2024-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 ..
 .. Redistribution and use in source and binary forms, with or without
 .. modification, are permitted provided that the following conditions
@@ -29,11 +29,10 @@
 Python
 ####
 
-.. include:: python_readme.rst
-
 .. toctree::
    :maxdepth: 1
    :hidden:
 
+   Overview <../tutorials/Triton_Inference_Server_Python_API/README.md>
    Kafka I/O <../tutorials/Triton_Inference_Server_Python_API/examples/kafka-io/README.md>
    Rayserve <../tutorials/Triton_Inference_Server_Python_API/examples/rayserve/README.md>
\ No newline at end of file
diff --git a/docs/client_guide/python_readme.rst b/docs/client_guide/python_readme.rst
deleted file mode 100644
index cbcf5c9b25..0000000000
--- a/docs/client_guide/python_readme.rst
+++ /dev/null
@@ -1,300 +0,0 @@
-..
-.. Copyright 2024-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
-..
-.. Redistribution and use in source and binary forms, with or without
-.. modification, are permitted provided that the following conditions
-.. are met:
-..  * Redistributions of source code must retain the above copyright
-..    notice, this list of conditions and the following disclaimer.
-..  * Redistributions in binary form must reproduce the above copyright
-..    notice, this list of conditions and the following disclaimer in the
-..    documentation and/or other materials provided with the distribution.
-..  * Neither the name of NVIDIA CORPORATION nor the names of its
-..    contributors may be used to endorse or promote products derived
-..    from this software without specific prior written permission.
-..
-.. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
-.. EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-.. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-.. PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-.. CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-.. EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-.. PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-.. PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
-.. OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-.. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-.. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-.. raw:: html
-
-
-Triton Inference Server In-Process Python API [BETA]
-====================================================
-
-Starting with release 24.01 Triton Inference Server will include a
-Python package enabling developers to embed Triton Inference Server
-instances in their Python applications. The in-process Python API is
-designed to match the functionality of the in-process C API while
-providing a higher level abstraction. At its core the API relies on a
-1:1 python binding of the C API and provides all the flexibility and
-power of the C API with a simpler to use interface.
-
-   [!Note] As the API is in BETA please expect some changes as we test
-   out different features and get feedback. All feedback is weclome and
-   we look forward to hearing from you!
-
-| `Requirements <#requirements>`__ \| `Installation <#installation>`__
-  \| `Hello World <#hello-world>`__ \| `Stable
-  Diffusion <#stable-diffusion>`__ \| `Ray Serve
-  Deployment <../tutorials/Triton_Inference_Server_Python_API/examples/rayserve>`__ \|
-Requirements
-------------
-
-The following instructions require a linux system with Docker installed.
-For CUDA support, make sure your CUDA driver meets the requirements in
-“NVIDIA Driver” section of Deep Learning Framework support matrix:
-https://docs.nvidia.com/deeplearning/frameworks/support-matrix/index.html
-
-Installation
-------------
-
-The tutorial and Python API package are designed to be installed and run
-within the ``nvcr.io/nvidia/tritonserver:24.01-py3`` docker image.
-
-A set of convenience scripts are provided to create a docker image based
-on the ``nvcr.io/nvidia/tritonserver:24.01-py3`` image with the Python
-API installed plus additional dependencies required for the examples.
-
-Triton Inference Server 24.01 + Python API
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Clone Repository
-^^^^^^^^^^^^^^^^
-
-.. code:: bash
-
-   git clone https://github.com/triton-inference-server/tutorials.git
-   cd tutorials/Triton_Inference_Server_Python_API
-
-Build ``triton-python-api:r24.01`` Image
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-.. code:: bash
-
-   ./build.sh
-
-Supported Backends
-^^^^^^^^^^^^^^^^^^
-
-The built image includes all the backends shipped by default in the
-tritonserver ``nvcr.io/nvidia/tritonserver:24.01-py3`` container.
-
-::
-
-   dali  fil  identity  onnxruntime  openvino  python  pytorch  repeat  square  tensorrt
-
-Included Models
-^^^^^^^^^^^^^^^
-
-The ``default`` build includes an ``identity`` model that can be used
-for exercising basic operations including sending input tensors of
-different data types. The ``identity`` model copies provided inputs of
-``shape [-1, -1]`` to outputs of shape ``[-1, -1]``. Inputs are named
-``data_type_input`` and outputs are named ``data_type_output``
-(e.g. ``string_input``, ``string_output``, ``fp16_input``,
-``fp16_output``).
-
-Hello World
------------
-
-Start ``triton-python-api:r24.01`` Container
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-The following command starts a container and volume mounts the current
-directory as ``workspace``.
-
-.. code:: bash
-
-   ./run.sh
-
-Enter Python Shell
-~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   python3
-
-Create and Start a Server Instance
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: python
-
-   import tritonserver
-   server = tritonserver.Server(model_repository="/workspace/identity-models")
-   server.start()
-
-List Models
-~~~~~~~~~~~
-
-::
-
-   server.models()
-
-Example Output
-^^^^^^^^^^^^^^
-
-``server.models()`` returns a dictionary of the available models with
-their current state.
-
-.. code:: python
-
-   {('identity', 1): {'name': 'identity', 'version': 1, 'state': 'READY'}}
-
-Send an Inference Request
-~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: python
-
-   model = server.model("identity")
-   responses = model.infer(inputs={"string_input":[["hello world!"]]})
-
-Iterate through Responses
-~~~~~~~~~~~~~~~~~~~~~~~~~
-
-``model.infer()`` returns an iterator that can be used to process the
-results of an inference request.
-
-.. code:: python
-
-   for response in responses:
-       print(response.outputs["string_output"].to_string_array())
-
-.. _example-output-1:
-
-Example Output
-^^^^^^^^^^^^^^
-
-.. code:: python
-
-   [['hello world!']]
-
-Stable Diffusion
-----------------
-
-This example is based on the
-`Popular_Models_Guide/StableDiffusion <../tutorials/Popular_Models_Guide/StableDiffusion/README.html>`__
-tutorial.
-
-Build ``triton-python-api:r24.01-diffusion`` Image and Stable Diffusion Models
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-Please note the following command will take many minutes depending on
-your hardware configuration and network connection.
-
-.. code:: bash
-
-   ./build.sh --framework diffusion --build-models
-
-.. _supported-backends-1:
-
-Supported Backends
-^^^^^^^^^^^^^^^^^^
-
-The built image includes all the backends shipped by default in the
-tritonserver ``nvcr.io/nvidia/tritonserver:24.01-py3`` container.
-
-::
-
-   dali  fil  identity  onnxruntime  openvino  python  pytorch  repeat  square  tensorrt
-
-.. _included-models-1:
-
-Included Models
-^^^^^^^^^^^^^^^
-
-The ``diffusion`` build includes a ``stable_diffustion`` pipeline that
-takes a text prompt and returns a generated image. For more details on
-the models and pipeline please see the
-`Popular_Models_Guide/StableDiffusion <../tutorials/Popular_Models_Guide/StableDiffusion/README.html>`__
-tutorial.
-
-Start Container
-~~~~~~~~~~~~~~~
-
-The following command starts a container and volume mounts the current
-directory as ``workspace``.
-
-.. code:: bash
-
-   ./run.sh --framework diffusion
-
-.. _enter-python-shell-1:
-
-Enter Python Shell
-~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   python3
-
-.. _create-and-start-a-server-instance-1:
-
-Create and Start a Server Instance
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: python
-
-   import tritonserver
-   import numpy
-   from PIL import Image
-   server = tritonserver.Server(model_repository="/workspace/diffusion-models")
-   server.start()
-
-.. _list-models-1:
-
-List Models
-~~~~~~~~~~~
-
-::
-
-   server.models()
-
-.. _example-output-2:
-
-Example Output
-^^^^^^^^^^^^^^
-
-.. code:: python
-
-   {('stable_diffusion', 1): {'name': 'stable_diffusion', 'version': 1, 'state': 'READY'}, ('text_encoder', 1): {'name': 'text_encoder', 'version': 1, 'state': 'READY'}, ('vae', 1): {'name': 'vae', 'version': 1, 'state': 'READY'}}
-
-.. _send-an-inference-request-1:
-
-Send an Inference Request
-~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: python
-
-   model = server.model("stable_diffusion")
-   responses = model.infer(inputs={"prompt":[["butterfly in new york, realistic, 4k, photograph"]]})
-
-Iterate through Responses and save image
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: python
-
-   for response in responses:
-       generated_image = numpy.from_dlpack(response.outputs["generated_image"])
-       generated_image = generated_image.squeeze().astype(numpy.uint8)
-       image_ = Image.fromarray(generated_image)
-       image_.save("sample_generated_image.jpg")
-
-.. _example-output-3:
-
-Example Output
-^^^^^^^^^^^^^^
-
-.. figure:: ../tutorials/Triton_Inference_Server_Python_API/docs/sample_generated_image.jpg
-   :alt: sample_generated_image
-
-   sample_generated_image
\ No newline at end of file
diff --git a/docs/contents.rst b/docs/contents.rst
index ca7ef3977f..420293aa76 100644
--- a/docs/contents.rst
+++ b/docs/contents.rst
@@ -36,7 +36,7 @@
    :hidden:
    :caption: Getting Started
 
-   getting_started/quick_deployment_by_backend
+   getting_started/quick_deployment
    LLM With TensorRT-LLM <getting_started/trtllm_user_guide.md>
    Multimodal model <../tutorials/Popular_Models_Guide/Llava1.5/llava_trtllm_guide.md>
    Stable diffusion <../tutorials/Popular_Models_Guide/StableDiffusion/README.md>
@@ -54,7 +54,7 @@
 
    Constrained Decoding <../tutorials/Feature_Guide/Constrained_Decoding/README.md>
    Function Calling <../tutorials/Feature_Guide/Function_Calling/README.md>
-   llm_features/speculative_decoding_by_backend_type
+   llm_features/speculative_decoding
 
 .. toctree::
    :hidden:
@@ -108,7 +108,7 @@
 
 .. toctree::
    :hidden:
-   :caption: Perf benchmarking and tuning
+   :caption: Performance benchmarking and tuning
 
    GenAI Perf Analyzer <perf_benchmark/genai_perf>
    Performance Analyzer <perf_benchmark/perf_analyzer>
diff --git a/docs/getting_started/quick_deployment_by_backend.rst b/docs/getting_started/quick_deployment.rst
similarity index 95%
rename from docs/getting_started/quick_deployment_by_backend.rst
rename to docs/getting_started/quick_deployment.rst
index 32f2708430..b20775684a 100644
--- a/docs/getting_started/quick_deployment_by_backend.rst
+++ b/docs/getting_started/quick_deployment.rst
@@ -1,5 +1,5 @@
 ..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+.. Copyright 2024-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 ..
 .. Redistribution and use in source and binary forms, with or without
 .. modification, are permitted provided that the following conditions
@@ -29,12 +29,11 @@
 Quick Deployment Guide by backend
 ####
 
-.. include:: quick_start.rst
-
 .. toctree::
    :maxdepth: 1
    :hidden:
 
+   Quickstart <quickstart.md>
    TRT-LLM <llm.md>
    vLLM <../tutorials/Popular_Models_Guide/Llama2/vllm_guide.md>
    Python with HuggingFace <../tutorials/Quick_Deploy/HuggingFaceTransformers/README.md>
diff --git a/docs/getting_started/quick_start.rst b/docs/getting_started/quick_start.rst
deleted file mode 100644
index 27f100e3cd..0000000000
--- a/docs/getting_started/quick_start.rst
+++ /dev/null
@@ -1,175 +0,0 @@
-..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
-..
-.. Redistribution and use in source and binary forms, with or without
-.. modification, are permitted provided that the following conditions
-.. are met:
-..  * Redistributions of source code must retain the above copyright
-..    notice, this list of conditions and the following disclaimer.
-..  * Redistributions in binary form must reproduce the above copyright
-..    notice, this list of conditions and the following disclaimer in the
-..    documentation and/or other materials provided with the distribution.
-..  * Neither the name of NVIDIA CORPORATION nor the names of its
-..    contributors may be used to endorse or promote products derived
-..    from this software without specific prior written permission.
-..
-.. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
-.. EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-.. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-.. PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-.. CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-.. EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-.. PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-.. PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
-.. OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-.. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-.. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-.. raw:: html
-
-
-Quickstart
-==========
-
-**New to Triton Inference Server and want do just deploy your model
-quickly?** Make use of `these
-tutorials <../tutorials/README.html#quick-deploy>`__ to begin your Triton
-journey!
-
-The Triton Inference Server is available as `buildable source
-code <../customization_guide/build.html>`__, but the easiest way to
-install and run Triton is to use the pre-built Docker image available
-from the `NVIDIA GPU Cloud (NGC) <https://ngc.nvidia.com>`__.
-
-Launching and maintaining Triton Inference Server revolves around the
-use of building model repositories. This tutorial will cover:
-
--  Creating a Model Repository
--  Launching Triton
--  Send an Inference Request
-
-Create A Model Repository
--------------------------
-
-The `model repository <../user_guide/model_repository.html>`__ is the
-directory where you place the models that you want Triton to serve. An
-example model repository is included in the
-`docs/examples/model_repository <https://github.com/triton-inference-server/server/blob/main/docs/examples/model_repository>`__.
-Before using the repository, you must fetch any missing model definition
-files from their public model zoos via the provided script.
-
-::
-
-   $ cd docs/examples
-   $ ./fetch_models.sh
-
-Launch Triton
--------------
-
-Triton is optimized to provide the best inferencing performance by using
-GPUs, but it can also work on CPU-only systems. In both cases you can
-use the same Triton Docker image.
-
-Run on System with GPUs
-~~~~~~~~~~~~~~~~~~~~~~~
-
-Use the following command to run Triton with the example model
-repository you just created. The `NVIDIA Container
-Toolkit <https://github.com/NVIDIA/nvidia-docker>`__ must be installed
-for Docker to recognize the GPU(s). The –gpus=1 flag indicates that 1
-system GPU should be made available to Triton for inferencing.
-
-::
-
-   $ docker run --gpus=1 --rm -p8000:8000 -p8001:8001 -p8002:8002 -v/full/path/to/docs/examples/model_repository:/models nvcr.io/nvidia/tritonserver:<xx.yy>-py3 tritonserver --model-repository=/models
-
-Where <xx.yy> is the version of Triton that you want to use (and pulled
-above). After you start Triton you will see output on the console
-showing the server starting up and loading the model. When you see
-output like the following, Triton is ready to accept inference requests.
-
-::
-
-   +----------------------+---------+--------+
-   | Model                | Version | Status |
-   +----------------------+---------+--------+
-   | <model_name>         | <v>     | READY  |
-   | ..                   | .       | ..     |
-   | ..                   | .       | ..     |
-   +----------------------+---------+--------+
-   ...
-   ...
-   ...
-   I1002 21:58:57.891440 62 grpc_server.cc:3914] Started GRPCInferenceService at 0.0.0.0:8001
-   I1002 21:58:57.893177 62 http_server.cc:2717] Started HTTPService at 0.0.0.0:8000
-   I1002 21:58:57.935518 62 http_server.cc:2736] Started Metrics Service at 0.0.0.0:8002
-
-All the models should show “READY” status to indicate that they loaded
-correctly. If a model fails to load the status will report the failure
-and a reason for the failure. If your model is not displayed in the
-table check the path to the model repository and your CUDA drivers.
-
-Run on CPU-Only System
-~~~~~~~~~~~~~~~~~~~~~~
-
-On a system without GPUs, Triton should be run without using the –gpus
-flag to Docker, but is otherwise identical to what is described above.
-
-::
-
-   $ docker run --rm -p8000:8000 -p8001:8001 -p8002:8002 -v/full/path/to/docs/examples/model_repository:/models nvcr.io/nvidia/tritonserver:<xx.yy>-py3 tritonserver --model-repository=/models
-
-Because the –gpus flag is not used, a GPU is not available and Triton
-will therefore be unable to load any model configuration that requires a
-GPU.
-
-Verify Triton Is Running Correctly
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Use Triton’s *ready* endpoint to verify that the server and the models
-are ready for inference. From the host system use curl to access the
-HTTP endpoint that indicates server status.
-
-::
-
-   $ curl -v localhost:8000/v2/health/ready
-   ...
-   < HTTP/1.1 200 OK
-   < Content-Length: 0
-   < Content-Type: text/plain
-
-The HTTP request returns status 200 if Triton is ready and non-200 if it
-is not ready.
-
-Send an Inference Request
--------------------------
-
-Use docker pull to get the client libraries and examples image from NGC.
-
-::
-
-   $ docker pull nvcr.io/nvidia/tritonserver:<xx.yy>-py3-sdk
-
-Where <xx.yy> is the version that you want to pull. Run the client
-image.
-
-::
-
-   $ docker run -it --rm --net=host nvcr.io/nvidia/tritonserver:<xx.yy>-py3-sdk
-
-From within the nvcr.io/nvidia/tritonserver:<xx.yy>-py3-sdk image, run
-the example image-client application to perform image classification
-using the example densenet_onnx model.
-
-To send a request for the densenet_onnx model use an image from the
-/workspace/images directory. In this case we ask for the top 3
-classifications.
-
-::
-
-   $ /workspace/install/bin/image_client -m densenet_onnx -c 3 -s INCEPTION /workspace/images/mug.jpg
-   Request 0, batch size 1
-   Image '/workspace/images/mug.jpg':
-       15.346230 (504) = COFFEE MUG
-       13.224326 (968) = CUP
-       10.422965 (505) = COFFEEPOT
\ No newline at end of file
diff --git a/docs/llm_features/speculative_decoding.rst b/docs/llm_features/speculative_decoding.rst
index debbcf52ae..7e0df42bc5 100644
--- a/docs/llm_features/speculative_decoding.rst
+++ b/docs/llm_features/speculative_decoding.rst
@@ -1,5 +1,5 @@
 ..
-.. Copyright 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+.. Copyright 2025-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 ..
 .. Redistribution and use in source and binary forms, with or without
 .. modification, are permitted provided that the following conditions
@@ -25,30 +25,14 @@
 .. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 .. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
-.. raw:: html
+####
+Speculative Decoding
+####
 
+.. toctree::
+   :maxdepth: 1
+   :hidden:
 
-About Speculative Decoding
-=========================
-Speculative Decoding (also referred to as Speculative Sampling) is a set of techniques designed
-to allow generation of more than one token per forward pass iteration. This can lead to a reduction
-in the average per-token latency in situations where the GPU is underutilized due to small batch sizes.
-
-Speculative decoding involves predicting a sequence of future tokens, referred to as draft tokens,
-using a method that is substantially more efficient than repeatedly executing the target Large Language
-Model (LLM). These draft tokens are then collectively validated by processing them through the target LLM
-in a single forward pass. The underlying assumptions are twofold:
-
-1. processing multiple draft tokens concurrently will be as rapid as processing a single token
-2. multiple draft tokens will be validated successfully over the course of the full generation
-
-If the first assumption holds true, the latency of speculative decoding will no worse than the standard
-approach. If the second holds, output token generation advances by statistically more than one token per
-forward pass. The combination of both these allows speculative decoding to result in reduced latency.
-
-Performance Improvements
-========================
-It's important to note that the effectiveness of speculative decoding techniques is highly dependent
-on the specific task at hand. For instance, forecasting subsequent tokens in a code-completion scenario
-may prove simpler than generating a summary for an article. `Spec-Bench <https://sites.google.com/view/spec-bench>`__
-shows the performance of different speculative decoding approaches on different tasks.
\ No newline at end of file
+   Overview <../tutorials/Feature_Guide/Speculative_Decoding/README.md>
+   TRT-LLM <../tutorials/Feature_Guide/Speculative_Decoding/TRT-LLM/README.md>
+   vLLM <../tutorials/Feature_Guide/Speculative_Decoding/vLLM/README.md>
\ No newline at end of file
diff --git a/docs/llm_features/speculative_decoding_by_backend_type.rst b/docs/llm_features/speculative_decoding_by_backend_type.rst
deleted file mode 100644
index a61f625626..0000000000
--- a/docs/llm_features/speculative_decoding_by_backend_type.rst
+++ /dev/null
@@ -1,39 +0,0 @@
-..
-.. Copyright 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
-..
-.. Redistribution and use in source and binary forms, with or without
-.. modification, are permitted provided that the following conditions
-.. are met:
-..  * Redistributions of source code must retain the above copyright
-..    notice, this list of conditions and the following disclaimer.
-..  * Redistributions in binary form must reproduce the above copyright
-..    notice, this list of conditions and the following disclaimer in the
-..    documentation and/or other materials provided with the distribution.
-..  * Neither the name of NVIDIA CORPORATION nor the names of its
-..    contributors may be used to endorse or promote products derived
-..    from this software without specific prior written permission.
-..
-.. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
-.. EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-.. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-.. PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-.. CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-.. EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-.. PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-.. PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
-.. OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-.. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-.. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-####
-Speculative Decoding
-####
-
-.. include:: speculative_decoding.rst
-
-.. toctree::
-   :maxdepth: 1
-   :hidden:
-
-   TRT-LLM <../tutorials/Feature_Guide/Speculative_Decoding/TRT-LLM/README.md>
-   vLLM <../tutorials/Feature_Guide/Speculative_Decoding/vLLM/README.md>
\ No newline at end of file
diff --git a/docs/perf_benchmark/genai-perf-README.rst b/docs/perf_benchmark/genai-perf-README.rst
deleted file mode 100644
index c4a3c7d73d..0000000000
--- a/docs/perf_benchmark/genai-perf-README.rst
+++ /dev/null
@@ -1,686 +0,0 @@
-..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
-..
-.. Redistribution and use in source and binary forms, with or without
-.. modification, are permitted provided that the following conditions
-.. are met:
-..  * Redistributions of source code must retain the above copyright
-..    notice, this list of conditions and the following disclaimer.
-..  * Redistributions in binary form must reproduce the above copyright
-..    notice, this list of conditions and the following disclaimer in the
-..    documentation and/or other materials provided with the distribution.
-..  * Neither the name of NVIDIA CORPORATION nor the names of its
-..    contributors may be used to endorse or promote products derived
-..    from this software without specific prior written permission.
-..
-.. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
-.. EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-.. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-.. PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-.. CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-.. EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-.. PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-.. PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
-.. OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-.. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-.. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-.. raw:: html
-
-
-GenAI-Perf
-==========
-
-GenAI-Perf is a command line tool for measuring the throughput and
-latency of generative AI models as served through an inference server.
-For large language models (LLMs), GenAI-Perf provides metrics such as
-`output token throughput <#output_token_throughput_metric>`__, `time to
-first token <#time_to_first_token_metric>`__, `inter token
-latency <#inter_token_latency_metric>`__, and `request
-throughput <#request_throughput_metric>`__. For a full list of metrics
-please see the `Metrics section <#metrics>`__.
-
-Users specify a model name, an inference server URL, the type of inputs
-to use (synthetic or from dataset), and the type of load to generate
-(number of concurrent requests, request rate).
-
-GenAI-Perf generates the specified load, measures the performance of the
-inference server and reports the metrics in a simple table as console
-output. The tool also logs all results in a csv and json file that can
-be used to derive additional metrics and visualizations. The inference
-server must already be running when GenAI-Perf is run.
-
-You can use GenAI-Perf to run performance benchmarks on - `Large
-Language Models <docs/tutorial.md>`__ - `Vision Language
-Models <docs/multi_modal.md>`__ - `Embedding
-Models <docs/embeddings.md>`__ - `Ranking Models <docs/rankings.md>`__ -
-`Multiple LoRA Adapters <docs/lora.md>`__
-
-   [!Note] GenAI-Perf is currently in early release and under rapid
-   development. While we will try to remain consistent, command line
-   options and functionality are subject to change as the tool matures.
-
-.. raw:: html
-
-   <!--
-   ======================
-   INSTALLATION
-   ======================
-   -->
-
-Installation
-------------
-
-The easiest way to install GenAI-Perf is through `Triton Server SDK
-container <https://ngc.nvidia.com/catalog/containers/nvidia:tritonserver>`__.
-Install the latest release using the following command:
-
-.. code:: bash
-
-   export RELEASE="yy.mm" # e.g. export RELEASE="24.06"
-
-   docker run -it --net=host --gpus=all  nvcr.io/nvidia/tritonserver:${RELEASE}-py3-sdk
-
-   # Check out genai_perf command inside the container:
-   genai-perf --help
-
-.. raw:: html
-
-   <details>
-
-Alternatively, to install from source:
-
-Since GenAI-Perf depends on Perf Analyzer, you’ll need to install the
-Perf Analyzer binary:
-
-Install Perf Analyzer (Ubuntu, Python 3.8+)
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-**NOTE**: you must already have CUDA 12 installed (checkout the `CUDA
-installation
-guide <https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html>`__).
-
-.. code:: bash
-
-   pip install tritonclient
-
-   apt update && apt install -y --no-install-recommends libb64-0d libcurl4
-
-You can also build Perf Analyzer `from
-source <../docs/install.md#build-from-source>`__ as well.
-
-Install GenAI-Perf from source
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   git clone https://github.com/triton-inference-server/perf_analyzer.git && cd perf_analyzer
-
-   pip install -e genai-perf
-
-.. raw:: html
-
-   </details>
-
-.. raw:: html
-
-   <!--
-   ======================
-   QUICK START
-   ======================
-   -->
-
-Quick Start
------------
-
-In this quick start, we will use GenAI-Perf to run performance
-benchmarking on the GPT-2 model running on Triton Inference Server with
-a TensorRT-LLM engine.
-
-Serve GPT-2 TensorRT-LLM model using Triton CLI
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-You can follow the `quickstart
-guide <https://github.com/triton-inference-server/triton_cli?tab=readme-ov-file#serving-a-trt-llm-model>`__
-on Triton CLI github repo to run GPT-2 model locally. The full
-instructions are copied below for convenience:
-
-.. code:: bash
-
-   # This container comes with all of the dependencies for building TRT-LLM engines
-   # and serving the engine with Triton Inference Server.
-   docker run -ti \
-       --gpus all \
-       --network=host \
-       --shm-size=1g --ulimit memlock=-1 \
-       -v /tmp:/tmp \
-       -v ${HOME}/models:/root/models \
-       -v ${HOME}/.cache/huggingface:/root/.cache/huggingface \
-       nvcr.io/nvidia/tritonserver:24.05-trtllm-python-py3
-
-   # Install the Triton CLI
-   pip install git+https://github.com/triton-inference-server/triton_cli.git@0.0.8
-
-   # Build TRT LLM engine and generate a Triton model repository pointing at it
-   triton remove -m all
-   triton import -m gpt2 --backend tensorrtllm
-
-   # Start Triton pointing at the default model repository
-   triton start
-
-Running GenAI-Perf
-~~~~~~~~~~~~~~~~~~
-
-Now we can run GenAI-Perf from Triton Inference Server SDK container:
-
-.. code:: bash
-
-   export RELEASE="yy.mm" # e.g. export RELEASE="24.06"
-
-   docker run -it --net=host --rm --gpus=all nvcr.io/nvidia/tritonserver:${RELEASE}-py3-sdk
-
-   # Run GenAI-Perf in the container:
-   genai-perf profile \
-     -m gpt2 \
-     --service-kind triton \
-     --backend tensorrtllm \
-     --num-prompts 100 \
-     --random-seed 123 \
-     --synthetic-input-tokens-mean 200 \
-     --synthetic-input-tokens-stddev 0 \
-     --streaming \
-     --output-tokens-mean 100 \
-     --output-tokens-stddev 0 \
-     --output-tokens-mean-deterministic \
-     --tokenizer hf-internal-testing/llama-tokenizer \
-     --concurrency 1 \
-     --measurement-interval 4000 \
-     --profile-export-file my_profile_export.json \
-     --url localhost:8001
-
-Example output:
-
-::
-
-                                      LLM Metrics
-   ┏━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━━┳━━━━━━━━┳━━━━━━━━┳━━━━━━━━┳━━━━━━━━┓
-   ┃                Statistic ┃    avg ┃    min ┃    max ┃    p99 ┃    p90 ┃    p75 ┃
-   ┡━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━━╇━━━━━━━━╇━━━━━━━━╇━━━━━━━━╇━━━━━━━━┩
-   │ Time to first token (ms) │  11.70 │   9.88 │  17.21 │  14.35 │  12.01 │  11.87 │
-   │ Inter token latency (ms) │   1.46 │   1.08 │   1.89 │   1.87 │   1.62 │   1.52 │
-   │     Request latency (ms) │ 161.24 │ 153.45 │ 200.74 │ 200.66 │ 179.43 │ 162.23 │
-   │   Output sequence length │ 103.39 │  95.00 │ 134.00 │ 120.08 │ 107.30 │ 105.00 │
-   │    Input sequence length │ 200.01 │ 200.00 │ 201.00 │ 200.13 │ 200.00 │ 200.00 │
-   └──────────────────────────┴────────┴────────┴────────┴────────┴────────┴────────┘
-   Output token throughput (per sec): 635.61
-   Request throughput (per sec): 6.15
-
-See `Tutorial <docs/tutorial.md>`__ for additional examples.
-
-.. raw:: html
-
-   <!--
-   ======================
-   VISUALIZATION
-   ======================
-   -->
-
-Visualization
--------------
-
-GenAI-Perf can also generate various plots that visualize the
-performance of the current profile run. This is disabled by default but
-users can easily enable it by passing the ``--generate-plots`` option
-when running the benchmark:
-
-.. code:: bash
-
-   genai-perf profile \
-     -m gpt2 \
-     --service-kind triton \
-     --backend tensorrtllm \
-     --streaming \
-     --concurrency 1 \
-     --generate-plots
-
-This will generate a `set of default
-plots <docs/compare.md#example-plots>`__ such as: - Time to first token
-(TTFT) analysis - Request latency analysis - TTFT vs Input sequence
-lengths - Inter token latencies vs Token positions - Input sequence
-lengths vs Output sequence lengths
-
-Using ``compare`` Subcommand to Visualize Multiple Runs
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-The ``compare`` subcommand in GenAI-Perf facilitates users in comparing
-multiple profile runs and visualizing the differences through plots.
-
-Usage
-^^^^^
-
-Assuming the user possesses two profile export JSON files, namely
-``profile1.json`` and ``profile2.json``, they can execute the
-``compare`` subcommand using the ``--files`` option:
-
-.. code:: bash
-
-   genai-perf compare --files profile1.json profile2.json
-
-Executing the above command will perform the following actions under the
-``compare`` directory: 1. Generate a YAML configuration file
-(e.g. ``config.yaml``) containing the metadata for each plot generated
-during the comparison process. 2. Automatically generate the `default
-set of plots <docs/compare.md#example-plots>`__ (e.g. TTFT vs. Input
-Sequence Lengths) that compare the two profile runs.
-
-::
-
-   compare
-   ├── config.yaml
-   ├── distribution_of_input_sequence_lengths_to_output_sequence_lengths.jpeg
-   ├── request_latency.jpeg
-   ├── time_to_first_token.jpeg
-   ├── time_to_first_token_vs_input_sequence_lengths.jpeg
-   ├── token-to-token_latency_vs_output_token_position.jpeg
-   └── ...
-
-Customization
-^^^^^^^^^^^^^
-
-Users have the flexibility to iteratively modify the generated YAML
-configuration file to suit their specific requirements. They can make
-alterations to the plots according to their preferences and execute the
-command with the ``--config`` option followed by the path to the
-modified configuration file:
-
-.. code:: bash
-
-   genai-perf compare --config compare/config.yaml
-
-This command will regenerate the plots based on the updated
-configuration settings, enabling users to refine the visual
-representation of the comparison results as per their needs.
-
-See `Compare documentation <docs/compare.md>`__ for more details.
-
-.. raw:: html
-
-   <!--
-   ======================
-   MODEL INPUTS
-   ======================
-   -->
-
-Model Inputs
-------------
-
-GenAI-Perf supports model input prompts from either synthetically
-generated inputs, or from the HuggingFace
-`OpenOrca <https://huggingface.co/datasets/Open-Orca/OpenOrca>`__ or
-`CNN_DailyMail <https://huggingface.co/datasets/cnn_dailymail>`__
-datasets. This is specified using the ``--input-dataset`` CLI option.
-
-When the dataset is synthetic, you can specify the following options: \*
-``--num-prompts <int>``: The number of unique prompts to generate as
-stimulus, >= 1. \* ``--synthetic-input-tokens-mean <int>``: The mean of
-number of tokens in the generated prompts when using synthetic data, >=
-1. \* ``--synthetic-input-tokens-stddev <int>``: The standard deviation
-of number of tokens in the generated prompts when using synthetic data,
->= 0. \* ``--random-seed <int>``: The seed used to generate random
-values, >= 0.
-
-When the dataset is coming from HuggingFace, you can specify the
-following options: \* ``--input-dataset {openorca,cnn_dailymail}``:
-HuggingFace dataset to use for benchmarking. \* ``--num-prompts <int>``:
-The number of unique prompts to generate as stimulus, >= 1.
-
-When the dataset is coming from a file, you can specify the following
-options: \* ``--input-file <path>``: The input file containing the
-prompts to use for benchmarking as JSON objects.
-
-For any dataset, you can specify the following options: \*
-``--output-tokens-mean <int>``: The mean number of tokens in each
-output. Ensure the ``--tokenizer`` value is set correctly, >= 1. \*
-``--output-tokens-stddev <int>``: The standard deviation of the number
-of tokens in each output. This is only used when output-tokens-mean is
-provided, >= 1. \* ``--output-tokens-mean-deterministic``: When using
-``--output-tokens-mean``, this flag can be set to improve precision by
-setting the minimum number of tokens equal to the requested number of
-tokens. This is currently supported with the Triton service-kind. Note
-that there is still some variability in the requested number of output
-tokens, but GenAi-Perf attempts its best effort with your model to get
-the right number of output tokens.
-
-You can optionally set additional model inputs with the following
-option: \* ``--extra-inputs <input_name>:<value>``: An additional input
-for use with the model with a singular value, such as ``stream:true`` or
-``max_tokens:5``. This flag can be repeated to supply multiple extra
-inputs.
-
-For `Large Language Models <docs/tutorial.md>`__, there is no batch size
-(i.e. batch size is always ``1``). Each request includes the inputs for
-one individual inference. Other modes such as the
-`embeddings <docs/embeddings.md>`__ and `rankings <docs/rankings.md>`__
-endpoints support client-side batching, where ``--batch-size N`` means
-that each request sent will include the inputs for ``N`` separate
-inferences, allowing them to be processed together.
-
-.. raw:: html
-
-   <!--
-   ======================
-   METRICS
-   ======================
-   -->
-
-Metrics
--------
-
-GenAI-Perf collects a diverse set of metrics that captures the
-performance of the inference server.
-
-+-----------------------+-----------------------+-----------------------+
-| Metric                | Description           | Aggregations          |
-+=======================+=======================+=======================+
-| Time to First Token   | Time between when a   | Avg, min, max, p99,   |
-|                       | request is sent and   | p90, p75              |
-|                       | when its first        |                       |
-|                       | response is received, |                       |
-|                       | one value per request |                       |
-|                       | in benchmark          |                       |
-+-----------------------+-----------------------+-----------------------+
-| Inter Token Latency   | Time between          | Avg, min, max, p99,   |
-|                       | intermediate          | p90, p75              |
-|                       | responses for a       |                       |
-|                       | single request        |                       |
-|                       | divided by the number |                       |
-|                       | of generated tokens   |                       |
-|                       | of the latter         |                       |
-|                       | response, one value   |                       |
-|                       | per response per      |                       |
-|                       | request in benchmark  |                       |
-+-----------------------+-----------------------+-----------------------+
-| Request Latency       | Time between when a   | Avg, min, max, p99,   |
-|                       | request is sent and   | p90, p75              |
-|                       | when its final        |                       |
-|                       | response is received, |                       |
-|                       | one value per request |                       |
-|                       | in benchmark          |                       |
-+-----------------------+-----------------------+-----------------------+
-| Output Sequence       | Total number of       | Avg, min, max, p99,   |
-| Length                | output tokens of a    | p90, p75              |
-|                       | request, one value    |                       |
-|                       | per request in        |                       |
-|                       | benchmark             |                       |
-+-----------------------+-----------------------+-----------------------+
-| Input Sequence Length | Total number of input | Avg, min, max, p99,   |
-|                       | tokens of a request,  | p90, p75              |
-|                       | one value per request |                       |
-|                       | in benchmark          |                       |
-+-----------------------+-----------------------+-----------------------+
-| Output Token          | Total number of       | None–one value per    |
-| Throughput            | output tokens from    | benchmark             |
-|                       | benchmark divided by  |                       |
-|                       | benchmark duration    |                       |
-+-----------------------+-----------------------+-----------------------+
-| Request Throughput    | Number of final       | None–one value per    |
-|                       | responses from        | benchmark             |
-|                       | benchmark divided by  |                       |
-|                       | benchmark duration    |                       |
-+-----------------------+-----------------------+-----------------------+
-
-.. raw:: html
-
-   <!--
-   ======================
-   COMMAND LINE OPTIONS
-   ======================
-   -->
-
-Command Line Options
---------------------
-
-``-h``
-''''''
-
-``--help``
-''''''''''
-
-Show the help message and exit.
-
-Endpoint Options:
-~~~~~~~~~~~~~~~~~
-
-``-m <list>``
-'''''''''''''
-
-``--model <list>``
-''''''''''''''''''
-
-The names of the models to benchmark. A single model is recommended,
-unless you are `profiling multiple LoRA adapters <docs/lora.md>`__.
-(default: ``None``)
-
-``--model-selection-strategy {round_robin, random}``
-''''''''''''''''''''''''''''''''''''''''''''''''''''
-
-When multiple models are specified, this is how a specific model is
-assigned to a prompt. Round robin means that each model receives a
-request in order. Random means that assignment is uniformly random
-(default: ``round_robin``)
-
-``--backend {tensorrtllm,vllm}``
-''''''''''''''''''''''''''''''''
-
-When using the “triton” service-kind, this is the backend of the model.
-For the TRT-LLM backend, you currently must set
-``exclude_input_in_output`` to true in the model config to not echo the
-input tokens in the output. (default: tensorrtllm)
-
-``--endpoint <str>``
-''''''''''''''''''''
-
-Set a custom endpoint that differs from the OpenAI defaults. (default:
-``None``)
-
-``--endpoint-type {chat,completions,embeddings,rankings}``
-''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
-
-The endpoint-type to send requests to on the server. This is only used
-with the ``openai`` service-kind. (default: ``None``)
-
-``--service-kind {triton,openai}``
-''''''''''''''''''''''''''''''''''
-
-The kind of service perf_analyzer will generate load for. In order to
-use ``openai``, you must specify an api via ``--endpoint-type``.
-(default: ``triton``)
-
-``--streaming``
-'''''''''''''''
-
-An option to enable the use of the streaming API. (default: ``False``)
-
-``-u <url>``
-''''''''''''
-
-``--url <url>``
-'''''''''''''''
-
-URL of the endpoint to target for benchmarking. (default: ``None``)
-
-Input Options
-~~~~~~~~~~~~~
-
-``-b <int>``
-''''''''''''
-
-``--batch-size <int>``
-''''''''''''''''''''''
-
-The batch size of the requests GenAI-Perf should send. This is currently
-only supported with the `embeddings <docs/embeddings.md>`__,
-image_retrieval, and `rankings <docs/rankings.md>`__ endpoint types.
-(default: ``1``)
-
-``--extra-inputs <str>``
-''''''''''''''''''''''''
-
-Provide additional inputs to include with every request. You can repeat
-this flag for multiple inputs. Inputs should be in an input_name:value
-format. Alternatively, a string representing a json formatted dict can
-be provided. (default: ``None``)
-
-``--input-dataset {openorca,cnn_dailymail}``
-''''''''''''''''''''''''''''''''''''''''''''
-
-The HuggingFace dataset to use for prompts. (default: ``openorca``)
-
-``--input-file <path>``
-'''''''''''''''''''''''
-
-The input file containing the prompts to use for profiling. Each line
-should be a JSON object with a ‘text_input’ field in JSONL format.
-Example: {"text_input": "Your prompt here"}"
-
-``--num-prompts <int>``
-'''''''''''''''''''''''
-
-The number of unique prompts to generate as stimulus. (default: ``100``)
-
-``--output-tokens-mean <int>``
-''''''''''''''''''''''''''''''
-
-The mean number of tokens in each output. Ensure the ``--tokenizer``
-value is set correctly. (default: ``-1``)
-
-``--output-tokens-mean-deterministic``
-''''''''''''''''''''''''''''''''''''''
-
-When using ``--output-tokens-mean``, this flag can be set to improve
-precision by setting the minimum number of tokens equal to the requested
-number of tokens. This is currently supported with the Triton
-service-kind. Note that there is still some variability in the requested
-number of output tokens, but GenAi-Perf attempts its best effort with
-your model to get the right number of output tokens. (default:
-``False``)
-
-``--output-tokens-stddev <int>``
-''''''''''''''''''''''''''''''''
-
-The standard deviation of the number of tokens in each output. This is
-only used when ``--output-tokens-mean`` is provided. (default: ``0``)
-
-``--random-seed <int>``
-'''''''''''''''''''''''
-
-The seed used to generate random values. (default: ``0``)
-
-``--synthetic-input-tokens-mean <int>``
-'''''''''''''''''''''''''''''''''''''''
-
-The mean of number of tokens in the generated prompts when using
-synthetic data. (default: ``550``)
-
-``--synthetic-input-tokens-stddev <int>``
-'''''''''''''''''''''''''''''''''''''''''
-
-The standard deviation of number of tokens in the generated prompts when
-using synthetic data. (default: ``0``)
-
-Profiling Options
-~~~~~~~~~~~~~~~~~
-
-``--concurrency <int>``
-'''''''''''''''''''''''
-
-The concurrency value to benchmark. (default: ``None``)
-
-``--measurement-interval <int>``
-''''''''''''''''''''''''''''''''
-
-``-p <int>``
-''''''''''''
-
-The time interval used for each measurement in milliseconds. Perf
-Analyzer will sample a time interval specified and take measurement over
-the requests completed within that time interval. (default: ``10000``)
-
-``--request-rate <float>``
-''''''''''''''''''''''''''
-
-Sets the request rate for the load generated by PA. (default: ``None``)
-
-``-s <float>``
-''''''''''''''
-
-``--stability-percentage <float>``
-''''''''''''''''''''''''''''''''''
-
-The allowed variation in latency measurements when determining if a
-result is stable. The measurement is considered as stable if the ratio
-of max / min from the recent 3 measurements is within (stability
-percentage) in terms of both infer per second and latency. (default:
-``999``)
-
-Output Options
-~~~~~~~~~~~~~~
-
-``--artifact-dir``
-''''''''''''''''''
-
-The directory to store all the (output) artifacts generated by
-GenAI-Perf and Perf Analyzer. (default: ``artifacts``)
-
-``--generate-plots``
-''''''''''''''''''''
-
-An option to enable the generation of plots. (default: False)
-
-``--profile-export-file <path>``
-''''''''''''''''''''''''''''''''
-
-The path where the perf_analyzer profile export will be generated. By
-default, the profile export will be to ``profile_export.json``. The
-genai-perf files will be exported to
-``<profile_export_file>_genai_perf.json`` and
-``<profile_export_file>_genai_perf.csv``. For example, if the profile
-export file is ``profile_export.json``, the genai-perf file will be
-exported to ``profile_export_genai_perf.csv``. (default:
-``profile_export.json``)
-
-Other Options
-~~~~~~~~~~~~~
-
-``--tokenizer <str>``
-'''''''''''''''''''''
-
-The HuggingFace tokenizer to use to interpret token metrics from prompts
-and responses. (default: ``hf-internal-testing/llama-tokenizer``)
-
-``-v``
-''''''
-
-``--verbose``
-'''''''''''''
-
-An option to enable verbose mode. (default: ``False``)
-
-``--version``
-'''''''''''''
-
-An option to print the version and exit.
-
-.. raw:: html
-
-   <!--
-   ======================
-   Known Issues
-   ======================
-   -->
-
-Known Issues
-------------
-
--  GenAI-Perf can be slow to finish if a high request-rate is provided
--  Token counts may not be exact
diff --git a/docs/perf_benchmark/genai_perf.rst b/docs/perf_benchmark/genai_perf.rst
index 175662477f..65de2d8ce1 100644
--- a/docs/perf_benchmark/genai_perf.rst
+++ b/docs/perf_benchmark/genai_perf.rst
@@ -1,5 +1,5 @@
 ..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+.. Copyright 2024-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 ..
 .. Redistribution and use in source and binary forms, with or without
 .. modification, are permitted provided that the following conditions
@@ -28,13 +28,12 @@
 ####
 GenAI Performance Analyzer
 ####
-.. include:: genai-perf-README.rst
-
 
 .. toctree::
    :maxdepth: 1
    :hidden:
 
+   Overview <../perf_analyzer/genai-perf/README.md>
    Large language models <../perf_analyzer/genai-perf/docs/tutorial.md>
    Visual language models <../perf_analyzer/genai-perf/docs/multi_modal.md>
    Embedding models <../perf_analyzer/genai-perf/docs/embeddings.md>
diff --git a/docs/perf_benchmark/model-analyzer-README.rst b/docs/perf_benchmark/model-analyzer-README.rst
deleted file mode 100644
index f31e7ca633..0000000000
--- a/docs/perf_benchmark/model-analyzer-README.rst
+++ /dev/null
@@ -1,203 +0,0 @@
-..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
-..
-.. Redistribution and use in source and binary forms, with or without
-.. modification, are permitted provided that the following conditions
-.. are met:
-..  * Redistributions of source code must retain the above copyright
-..    notice, this list of conditions and the following disclaimer.
-..  * Redistributions in binary form must reproduce the above copyright
-..    notice, this list of conditions and the following disclaimer in the
-..    documentation and/or other materials provided with the distribution.
-..  * Neither the name of NVIDIA CORPORATION nor the names of its
-..    contributors may be used to endorse or promote products derived
-..    from this software without specific prior written permission.
-..
-.. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
-.. EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-.. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-.. PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-.. CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-.. EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-.. PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-.. PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
-.. OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-.. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-.. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-.. raw:: html
-
-
-|License|
-
-Triton Model Analyzer
-=====================
-
-   [!Warning]
-
-   .. rubric:: LATEST RELEASE
-      :name: latest-release
-
-   You are currently on the ``main`` branch which tracks
-   under-development progress towards the next release. The latest
-   release of the Triton Model Analyzer is 1.42.0 and is available on
-   branch
-   `r24.07 <https://github.com/triton-inference-server/model_analyzer/tree/r24.07>`__.
-
-Triton Model Analyzer is a CLI tool which can help you find a more
-optimal configuration, on a given piece of hardware, for single,
-multiple, ensemble, or BLS models running on a `Triton Inference
-Server <https://github.com/triton-inference-server/server/>`__. Model
-Analyzer will also generate reports to help you better understand the
-trade-offs of the different configurations along with their compute and
-memory requirements.
-
-Features
-========
-
-Search Modes
-~~~~~~~~~~~~
-
--  `Optuna Search <docs/config_search.md#optuna-search-mode>`__ **-ALPHA
-   RELEASE-** allows you to search for every parameter that can be
-   specified in the model configuration, using a hyperparameter
-   optimization framework. Please see the
-   `Optuna <https://optuna.org/>`__ website if you are interested in
-   specific details on how the algorithm functions.
-
--  `Quick Search <docs/config_search.md#quick-search-mode>`__ will
-   **sparsely** search the `Max Batch
-   Size <https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#maximum-batch-size>`__,
-   `Dynamic
-   Batching <https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#dynamic-batcher>`__,
-   and `Instance
-   Group <https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#instance-groups>`__
-   spaces by utilizing a heuristic hill-climbing algorithm to help you
-   quickly find a more optimal configuration
-
--  `Automatic Brute
-   Search <docs/config_search.md#automatic-brute-search>`__ will
-   **exhaustively** search the `Max Batch
-   Size <https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#maximum-batch-size>`__,
-   `Dynamic
-   Batching <https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#dynamic-batcher>`__,
-   and `Instance
-   Group <https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#instance-groups>`__
-   parameters of your model configuration
-
--  `Manual Brute Search <docs/config_search.md#manual-brute-search>`__
-   allows you to create manual sweeps for every parameter that can be
-   specified in the model configuration
-
-Model Types
-~~~~~~~~~~~
-
--  `Ensemble <docs/model_types.md#ensemble>`__: Model Analyzer can help
-   you find the optimal settings when profiling an ensemble model
-
--  `BLS <docs/model_types.md#bls>`__: Model Analyzer can help you find
-   the optimal settings when profiling a BLS model
-
--  `Multi-Model <docs/model_types.md#multi-model>`__: Model Analyzer can
-   help you find the optimal settings when profiling multiple concurrent
-   models
-
--  `LLM <docs/model_types.md#llm>`__: Model Analyzer can help you find
-   the optimal settings when profiling Large Language Models
-
-Other Features
-~~~~~~~~~~~~~~
-
--  `Detailed and summary reports <docs/report.md>`__: Model Analyzer is
-   able to generate summarized and detailed reports that can help you
-   better understand the trade-offs between different model
-   configurations that can be used for your model.
-
--  `QoS Constraints <docs/config.md#constraint>`__: Constraints can help
-   you filter out the Model Analyzer results based on your QoS
-   requirements. For example, you can specify a latency budget to filter
-   out model configurations that do not satisfy the specified latency
-   threshold.
-
-Examples and Tutorials
-======================
-
-**Single Model**
-~~~~~~~~~~~~~~~~
-
-See the `Single Model Quick Start <docs/quick_start.md>`__ for a guide
-on how to use Model Analyzer to profile, analyze and report on a simple
-PyTorch model.
-
-**Multi Model**
-~~~~~~~~~~~~~~~
-
-See the `Multi-model Quick Start <docs/mm_quick_start.md>`__ for a guide
-on how to use Model Analyzer to profile, analyze and report on two
-models running concurrently on the same GPU.
-
-**Ensemble Model**
-~~~~~~~~~~~~~~~~~~
-
-See the `Ensemble Model Quick Start <docs/ensemble_quick_start.md>`__
-for a guide on how to use Model Analyzer to profile, analyze and report
-on a simple Ensemble model.
-
-**BLS Model**
-~~~~~~~~~~~~~
-
-See the `BLS Model Quick Start <docs/bls_quick_start.md>`__ for a guide
-on how to use Model Analyzer to profile, analyze and report on a simple
-BLS model.
-
-Documentation
-=============
-
--  `Installation <docs/install.md>`__
--  `Model Analyzer CLI <docs/cli.md>`__
--  `Launch Modes <docs/launch_modes.md>`__
--  `Configuring Model Analyzer <docs/config.md>`__
--  `Model Analyzer Metrics <docs/metrics.md>`__
--  `Model Config Search <docs/config_search.md>`__
--  `Model Types <docs/model_types.md>`__
--  `Checkpointing <docs/checkpoints.md>`__
--  `Model Analyzer Reports <docs/report.md>`__
--  `Deployment with Kubernetes <docs/kubernetes_deploy.md>`__
-
-Terminology
-===========
-
-Below are definitions of some commonly used terms in Model Analyzer:
-
--  **Model Type** - Category of model being profiled. Examples of this
-   include single, multi, ensemble, BLS, etc..
--  **Search Mode** - How Model Analyzer explores the possible
-   configuration space when profiling. This is either exhaustive (brute)
-   or heuristic (quick/optuna).
--  **Model Config Search** - The cross product of model type and search
-   mode.
--  **Launch Mode** - How the Triton Server is deployed and used by Model
-   Analyzer.
-
-Reporting problems, asking questions
-====================================
-
-We appreciate any feedback, questions or bug reporting regarding this
-project. When help with code is needed, follow the process outlined in
-the Stack Overflow (https://stackoverflow.com/help/mcve) document.
-Ensure posted examples are:
-
--  minimal – use as little code as possible that still produces the same
-   problem
-
--  complete – provide all parts needed to reproduce the problem. Check
-   if you can strip external dependency and still show the problem. The
-   less time we spend on reproducing problems the more time we have to
-   fix it
-
--  verifiable – test the code you’re about to provide to make sure it
-   reproduces the problem. Remove all other problems that are not
-   related to your request/question.
-
-.. |License| image:: https://img.shields.io/badge/License-Apache_2.0-lightgrey.svg
-   :target: https://opensource.org/licenses/Apache-2.0
diff --git a/docs/perf_benchmark/model_analyzer.rst b/docs/perf_benchmark/model_analyzer.rst
index c29a96aa92..a3c734ce30 100644
--- a/docs/perf_benchmark/model_analyzer.rst
+++ b/docs/perf_benchmark/model_analyzer.rst
@@ -1,5 +1,5 @@
 ..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+.. Copyright 2024-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 ..
 .. Redistribution and use in source and binary forms, with or without
 .. modification, are permitted provided that the following conditions
@@ -29,17 +29,23 @@
 Model Analyzer
 ####
 
-.. include:: model-analyzer-README.rst
-
 .. toctree::
    :maxdepth: 1
    :hidden:
 
-   ../model_analyzer/docs/cli.md
-   ../model_analyzer/docs/launch_modes.md
-   ../model_analyzer/docs/config.md
-   ../model_analyzer/docs/metrics.md
-   ../model_analyzer/docs/config_search.md
-   ../model_analyzer/docs/checkpoints.md
-   ../model_analyzer/docs/report.md
-   ../model_analyzer/docs/kubernetes_deploy.md
\ No newline at end of file
+   Overview <../model_analyzer/README.md>
+   Documentation <../model_analyzer/docs/README.md>
+   Quick Start <../model_analyzer/docs/quick_start.md>
+   Installation <../model_analyzer/docs/install.md>
+   CLI Reference <../model_analyzer/docs/cli.md>
+   Launch Modes <../model_analyzer/docs/launch_modes.md>
+   Configuration <../model_analyzer/docs/config.md>
+   Configuration Search <../model_analyzer/docs/config_search.md>
+   Metrics <../model_analyzer/docs/metrics.md>
+   Checkpointing <../model_analyzer/docs/checkpoints.md>
+   Reports <../model_analyzer/docs/report.md>
+   Kubernetes <../model_analyzer/docs/kubernetes_deploy.md>
+   Model Types <../model_analyzer/docs/model_types.md>
+   Ensemble Model <../model_analyzer/docs/ensemble_quick_start.md>
+   BLS Model <../model_analyzer/docs/bls_quick_start.md>
+   Multi-Model <../model_analyzer/docs/mm_quick_start.md>
\ No newline at end of file
diff --git a/docs/perf_benchmark/perf-analyzer-README.rst b/docs/perf_benchmark/perf-analyzer-README.rst
deleted file mode 100644
index df9c246eb3..0000000000
--- a/docs/perf_benchmark/perf-analyzer-README.rst
+++ /dev/null
@@ -1,175 +0,0 @@
-..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
-..
-.. Redistribution and use in source and binary forms, with or without
-.. modification, are permitted provided that the following conditions
-.. are met:
-..  * Redistributions of source code must retain the above copyright
-..    notice, this list of conditions and the following disclaimer.
-..  * Redistributions in binary form must reproduce the above copyright
-..    notice, this list of conditions and the following disclaimer in the
-..    documentation and/or other materials provided with the distribution.
-..  * Neither the name of NVIDIA CORPORATION nor the names of its
-..    contributors may be used to endorse or promote products derived
-..    from this software without specific prior written permission.
-..
-.. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
-.. EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-.. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-.. PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
-.. CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
-.. EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
-.. PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
-.. PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
-.. OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
-.. (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
-.. OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-.. raw:: html
-
-
-Triton Performance Analyzer
-===========================
-
-Triton Performance Analyzer is CLI tool which can help you optimize the
-inference performance of models running on Triton Inference Server by
-measuring changes in performance as you experiment with different
-optimization strategies.
-
-Features
-========
-
-Inference Load Modes
-~~~~~~~~~~~~~~~~~~~~
-
--  `Concurrency Mode <docs/inference_load_modes.md#concurrency-mode>`__
-   simlulates load by maintaining a specific concurrency of outgoing
-   requests to the server
-
--  `Request Rate
-   Mode <docs/inference_load_modes.md#request-rate-mode>`__ simulates
-   load by sending consecutive requests at a specific rate to the server
-
--  `Custom Interval
-   Mode <docs/inference_load_modes.md#custom-interval-mode>`__ simulates
-   load by sending consecutive requests at specific intervals to the
-   server
-
-Performance Measurement Modes
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
--  `Time Windows Mode <docs/measurements_metrics.md#time-windows>`__
-   measures model performance repeatedly over a specific time interval
-   until performance has stabilized
-
--  `Count Windows Mode <docs/measurements_metrics.md#count-windows>`__
-   measures model performance repeatedly over a specific number of
-   requests until performance has stabilized
-
-Other Features
-~~~~~~~~~~~~~~
-
--  `Sequence Models <../user_guide/architecture.md#stateful-models>`__,
-   `Ensemble Models <../user_guide/architecture.md#ensemble-models>`__,
-   and `Decoupled Models <../user_guide/decoupled_models.md>`__ can be
-   profiled in addition to standard/stateless/coupled models
-
--  `Input Data <docs/input_data.md>`__ to model inferences can be
-   auto-generated or specified as well as verifying output
-
-Quick Start
-===========
-
-The steps below will guide you on how to start using Perf Analyzer.
-
-Step 1: Start Triton Container
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   export RELEASE=<yy.mm> # e.g. to use the release from the end of February of 2023, do `export RELEASE=23.02`
-
-   docker pull nvcr.io/nvidia/tritonserver:${RELEASE}-py3
-
-   docker run --gpus all --rm -it --net host nvcr.io/nvidia/tritonserver:${RELEASE}-py3
-
-Step 2: Download ``simple`` Model
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   # inside triton container
-   git clone --depth 1 https://github.com/triton-inference-server/server
-
-   mkdir model_repository ; cp -r server/docs/examples/model_repository/simple model_repository
-
-Step 3: Start Triton Server
-~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   # inside triton container
-   tritonserver --model-repository $(pwd)/model_repository &> server.log &
-
-   # confirm server is ready, look for 'HTTP/1.1 200 OK'
-   curl -v localhost:8000/v2/health/ready
-
-   # detach (CTRL-p CTRL-q)
-
-Step 4: Start Triton SDK Container
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   docker pull nvcr.io/nvidia/tritonserver:${RELEASE}-py3-sdk
-
-   docker run --gpus all --rm -it --net host nvcr.io/nvidia/tritonserver:${RELEASE}-py3-sdk
-
-Step 5: Run Perf Analyzer
-~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. code:: bash
-
-   # inside sdk container
-   perf_analyzer -m simple
-
-See the full `quick start guide <docs/quick_start.md>`__ for additional
-tips on how to analyze output.
-
-Documentation
-=============
-
--  `Installation <docs/install.md>`__
--  `Perf Analyzer CLI <docs/cli.md>`__
--  `Inference Load Modes <docs/inference_load_modes.md>`__
--  `Input Data <docs/input_data.md>`__
--  `Measurements & Metrics <docs/measurements_metrics.md>`__
--  `Benchmarking <docs/benchmarking.md>`__
-
-Contributing
-============
-
-Contributions to Triton Perf Analyzer are more than welcome. To
-contribute please review the `contribution
-guidelines <https://github.com/triton-inference-server/server/blob/main/CONTRIBUTING.md>`__,
-then fork and create a pull request.
-
-Reporting problems, asking questions
-====================================
-
-We appreciate any feedback, questions or bug reporting regarding this
-project. When help with code is needed, follow the process outlined in
-the Stack Overflow (https://stackoverflow.com/help/mcve) document.
-Ensure posted examples are:
-
--  minimal - use as little code as possible that still produces the same
-   problem
-
--  complete - provide all parts needed to reproduce the problem. Check
-   if you can strip external dependency and still show the problem. The
-   less time we spend on reproducing problems the more time we have to
-   fix it
-
--  verifiable - test the code you’re about to provide to make sure it
-   reproduces the problem. Remove all other problems that are not
-   related to your request/question.
diff --git a/docs/perf_benchmark/perf_analyzer.rst b/docs/perf_benchmark/perf_analyzer.rst
index d6c6156a62..fcb25e0ca8 100644
--- a/docs/perf_benchmark/perf_analyzer.rst
+++ b/docs/perf_benchmark/perf_analyzer.rst
@@ -1,5 +1,5 @@
 ..
-.. Copyright 2024-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+.. Copyright 2024-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 ..
 .. Redistribution and use in source and binary forms, with or without
 .. modification, are permitted provided that the following conditions
@@ -29,14 +29,16 @@
 Performance Analyzer
 ####
 
-.. include:: perf-analyzer-README.rst
-
 .. toctree::
    :maxdepth: 1
    :hidden:
 
-   ../perf_analyzer/docs/install.md
-   ../perf_analyzer/docs/CLI.md
-   ../perf_analyzer/docs/inference_load_modes.md
-   ../perf_analyzer/docs/input_data.md
-   ../perf_analyzer/docs/measurements_metrics.md
\ No newline at end of file
+   Overview <../perf_analyzer/README.md>
+   Documentation <../perf_analyzer/docs/README.md>
+   Quick Start <../perf_analyzer/docs/quick_start.md>
+   Installation <../perf_analyzer/docs/install.md>
+   CLI Reference <../perf_analyzer/docs/cli.md>
+   Inference Load Modes <../perf_analyzer/docs/inference_load_modes.md>
+   Input Data <../perf_analyzer/docs/input_data.md>
+   Measurement Modes <../perf_analyzer/docs/measurements_metrics.md>
+   Benchmarking <../perf_analyzer/docs/benchmarking.md>
\ No newline at end of file