Qualcomm® AI Hub Apps

The Qualcomm® AI Hub Apps are a collection of sample machine learning apps ready to deploy on Qualcomm® devices.

Each app is designed to work with one or more models from Qualcomm® AI Hub Models.

With this repository, you can...

• Explore apps optimized for on-device deployment of various machine learning tasks.
• View open-source app recipes for running Qualcomm® AI Hub Models on local devices.

Supported runtimes

• TensorFlow Lite
• ONNX

Supported Deployment Targets

• Android 11 Red Velvet Cake & Newer, API v30+
• Windows 11

Supported compute units

• CPU, GPU, NPU (includes hexagon HTP)

Chipsets supported for NPU Acceleration

• Snapdragon 888/888+
• Snapdragon 8 Gen 1
• Snapdragon 8 Gen 2
• Snapdragon 8 Gen 3
• ... and all other Snapdragon® chipsets supported by the QNN SDK

Weight and activation type required for NPU Acceleration:

• Floating Point: FP16 (All Snapdragon® chipsets with Hexagon® Architecture v69 or newer)
• Integer : INT8 (All Snapdragon® chipsets)

NOTE: These apps will run without NPU acceleration on non-Snapdragon® chipsets.

Getting Started

1. Search for your desired OS & app in this folder, or in the app directory at the bottom of this file.

2. The README of the selected app will contain build & installation instructions.

Android App Directory

Task	Language	Inference API	Special Tags
Image Classification	Java	TensorFlow Lite
Semantic Segmentation	Java	TensorFlow Lite	OpenCV, Live Camera Feed
Super Resolution	Java	TensorFlow Lite

Windows App Directory

Task	Language	Inference API	Special Tags
Image Classification	C++	ONNX	OpenCV
Llama 2 Chat	C++	ONNX
Object Detection	C++	ONNX	OpenCV
Super Resolution	C++	ONNX	OpenCV
Whisper Speech-to-Text	Python	ONNX

LICENSE

Qualcomm® AI Hub Apps is licensed under BSD-3. See the LICENSE file.