Qualcomm AI Engine Direct
The Qualcomm AI Engine Direct SDK (commonly known as the QNN SDK (Qualcomm Neural Network SDK).) facilitates the use of various interfaces and artifacts across other processing units such as CPU, NPU, LPAI (Low Power AI), and so on, which are collectively known as backends. For more information, refer to Qualcomm documentation.
The Embedded Coder® Support Package for Qualcomm® Hexagon® Processors provides Simulink® block and MATLAB® System object™ to predict the backend's response using QNN host and target models, and verify the accuracy and performance before deploying to the backend.
Functions
qnn.HTP | Interface to predict responses of deep learning model for QNN HTP backend (Since R2025b) |
qnn.CPU | Interface to predict responses of deep learning model for QNN CPU backend (Since R2025b) |
qnn.LPAI | Interface to predict responses of deep learning model for QNN LPAI backend (Since R2025b) |
qnn.getNetworkProperties | Read values for properties defined in QNN model or QNN context binary file (Since R2025b) |
Blocks
| QNN HTP Predict | Predict responses of a QNN model or QNN context binary for the HTP (NPU) backend (Since R2025b) |
| QNN CPU Predict | Predict responses of a QNN model for the CPU backend (Since R2025b) |
| QNN LPAI Predict | Predict responses of a QNN context binary for the LPAI backend (Since R2025b) |
Featured Examples
Deploy Deep Neural Networks to Qualcomm Targets Using Qualcomm AI Engine Direct (QNN)
The complete workflow for using the Simulink® blocks under the QNN Simulink® library in Embedded Coder Support Package for Qualcomm® Hexagon® Processors. The workflow covers both simulation of deep learning networks on the host computer and Processor-in-the-Loop (PIL) validation on the target device.
