# AI/ML reference apps

The AI/ML reference applications showcase practical scenarios for
running models on a live camera feed, file source, or real-time
streaming protocol (RTSP) on Qualcomm Linux Development Kits. To try
out these AI reference applications, you must first obtain a model
from one of the mentioned model repositories, such as AI Hub or
TFLite model zoo. The subsequent steps are detailed below, including
the steps to run the reference applications.

The Qualcomm IM SDK contains sample applications that enable you to
experience the features of the Qualcomm Linux Development
Kit. These applications serve as a reference for custom use cases on
the Qualcomm Linux platform. See [Qualcomm IM SDK sample applications](https://docs.qualcomm.com/bundle/publicresource/topics/80-70017-50/example-applications.html)
for the list of default reference apps.

You can run the default reference application or customize the sample
application by integrating your own model.

Prerequisites

The prerequisites for running the reference applications are as follows:

- A Qualcomm Linux Development Kit: Connected to a monitor via
HDMI/DisplayPort
- An SSH connection with the Qualcomm Linux Development Kit; follow the
steps [here](https://docs.qualcomm.com/bundle/publicresource/topics/80-70017-254/how_to.html#use-ssh)
to establish an SSH connection with the device. Once connected, the
Qualcomm Linux Development Kit is accessible through its configured
IP address. The IP address is needed further to push models and
artifacts needed to run AI/ML applications.
- Model downloaded from AI Hub and the corresponding
[label](https://github.com/quic/sample-apps-for-qualcomm-linux/releases/download/GA1.3-labels/GA1.3-labels.zip)
files.
- Ensure that the device is powered on and the Wayland display is visible on the monitor.

Setup eSDK for development

The eSDK (extensible SDK) needs to be setup to develop
application/plugin code. See
[Compile application: Qualcomm IM SDK](https://docs.qualcomm.com/doc/80-70017-15B/topic/compile-application-qualcomm-im-sdk.html)
for instructions on setting up the eSDK and downloading and compiling the
source code.

- [Default reference application](https://docs.qualcomm.com/doc/80-70017-15B/topic/default-reference-app.html)
The *gst-ai-classification* sample application demonstrates the
capability of hardware to perform classification on a video stream.

The pipeline receives the video stream from the camera, file source,
or real-time streaming protocol (RTSP), performs preprocessing,
conducts inference on AI hardware, and displays the results on the
screen.
- [Customize reference application](https://docs.qualcomm.com/doc/80-70017-15B/topic/customize-reference-app.html)
It uses a custom-trained YoloV8 model as an example.

Last Published: Jan 21, 2026

[Previous Topic
Run models](https://docs.qualcomm.com/bundle/publicresource/80-70017-15B/topics/run-models.md) [Next Topic
Default reference application](https://docs.qualcomm.com/bundle/publicresource/80-70017-15B/topics/default-reference-app.md)