# Fine tune an AI model with custom data using Edge Impulse

Edge Impulse is an end-to-end platform for building edge AI models on Qualcomm®
Dragonwing™ devices.

It enables dataset creation, model training, and hardware-accelerated deployment.
The platform supports audio, image, sensor data, and custom models in multiple formats.

## Prerequisites

- Ensure that you have set up the device for the your development board.
- Ensure that you have established an SSH connection with the Qualcomm evaluation kit. Once connected, the evaluation kit is accessible through its configured IP address. The IP address is required to push models and artifacts for running AI/ML applications.

## Train an AI model

To start building models using Edge Impulse, do the following:

1. Sign up for a free developer account at [studio.edgeimpulse.com](https://studio.edgeimpulse.com).
2. Install the Edge Impulse command-line interface and other dependencies.

wget https://cdn.edgeimpulse.com/firmware/linux/setup-edge-impulse-qc-linux.sh
        Copy to clipboard

sh setup-edge-impulse-qc-linux.sh
        Copy to clipboard

source ~/.profile
        Copy to clipboard
3. Connect to Edge Impulse.

edge-impulse-linux
        Copy to clipboard

    The wizard launches and prompts you to log in and select an Edge Impulse project.

Tip

If you need to change to a different project or use an alternative camera, rerun the command with the <cite>–clean</cite> argument.
4. To confirm that your device is connected to Edge Impulse:

    1. Navigate to your Edge Impulse project and select Devices.
    2. Verify that your device appears in the list as shown in the following image.

![../_images/ai-edgeimpulse-connected-devices.png](data:image/png;base64,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)
5. To build your first AI model, follow one of the [Tutorials](https://docs.edgeimpulse.com/tutorials).
6. To run your model, from the terminal or the SSH session on your development board run the following command:

edge-impulse-linux-runner
        Copy to clipboard

    - This command will automatically build and download your model, and run it on the NPU. This command can be used only for the quantized models.
    - To manually download the EIM file, in the Deployment page of your Edge Impulse project, search for <cite>Linux (AARCH64 with Qualcomm QNN)</cite>.

## Bring your own model

Edge Impulse also supports bring your own model (BYOM) in formats such as SavedModel,
ONNX, LiteRT, TensorRT, or scikit-learn. Models deployed through BYOM are fully supported
on Qualcomm Dragonwing platforms, with NPU acceleration available for quantized models.

If your model isn’t already quantized, you can upload a representative dataset for better
performance.

To bring your own model for Qualcomm Dragonwing devices, see
[Edge Impulse &gt; Bring Your Own Model](https://docs.edgeimpulse.com/studio/projects/dashboard/byom).

Last Published: May 14, 2026

[Previous Topic
Run models](https://docs.qualcomm.com/bundle/publicresource/80-80022-15B/topics/run-models.md) [Next Topic
Fine tune model accuracy with AIMET](https://docs.qualcomm.com/bundle/publicresource/80-80022-15B/topics/aimet.md)