> ## Documentation Index
> Fetch the complete documentation index at: https://dragonwingdocs.qualcomm.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Use Qualcomm AI Runtime SDK to optimize an AI model

> Use the Qualcomm AI Runtime (QAIRT) SDK to convert, quantize, and run AI models.

Qualcomm AI Runtime SDK is an all-in-one SDK to port ML models to
run on Qualcomm hardware accelerators. The SDK contains
Qualcomm Neural Processing Engine (also known as SNPE) and
AI Engine Direct (also known as QNN) which offer tools to convert
and quantize models trained in PyTorch and TensorFlow and runtimes
to run these models on CPU, GPU, and HTP. Learn more about
[SNPE](https://docs.qualcomm.com/doc/80-63442-10/topic/index_SNPE.html)
and
[QNN](https://docs.qualcomm.com/doc/80-63442-10/topic/index_QNN.html).

<img src="https://mintcdn.com/qualcomm-prod/WwC9kmcnKl9Ef7de/Key-Documents/AI-Developer-Workflow/_images/optimize-compile-snpe.png?fit=max&auto=format&n=WwC9kmcnKl9Ef7de&q=85&s=509e572bace7c9b1c3ea3ad37edbade8" alt="QAIRT workflow" width="961" height="950" data-path="Key-Documents/AI-Developer-Workflow/_images/optimize-compile-snpe.png" />

## Setup

<CardGroup cols={2}>
  <Card title="Download SDK" icon="download" href="../topic/qairt-install">
    Download and install the Qualcomm AI Runtime SDK.
  </Card>

  <Card title="Setup SDK" icon="gear" href="../topic/qairt-setup">
    Configure the SDK environment for model optimization.
  </Card>
</CardGroup>

## Convert and quantize models

<CardGroup cols={2}>
  <Card title="Convert and quantize models" icon="arrows-spin" href="../topic/port-models">
    Convert models from PyTorch, TensorFlow, ONNX, or LiteRT and quantize them for deployment using SNPE or QNN.
  </Card>
</CardGroup>

## Run inference

<CardGroup cols={2}>
  <Card title="Run models" icon="play" href="../topic/run-models">
    Run inference on converted models using SNPE or QNN on CPU, GPU, or HTP.
  </Card>
</CardGroup>

## API reference

<CardGroup cols={2}>
  <Card title="Qualcomm Neural Processing Engine C/C++" icon="file-code" href="https://docs.qualcomm.com/doc/80-63442-10/topic/SNPE_general_api.html">
    SNPE C/C++ API reference documentation.
  </Card>

  <Card title="Qualcomm AI Engine Direct C/C++" icon="file-code" href="https://docs.qualcomm.com/doc/80-63442-10/topic/general_api.html">
    QNN C/C++ API reference documentation.
  </Card>
</CardGroup>