# Run machine learning use cases

LiteRT, Qualcomm Neural Processing SDK runtime, and open neural network exchange (ONNX) models are used for inference in the machine learning use cases.

Before you run the use cases, do the following:

- Complete the preconditions mentioned in [GStreamer command-line use cases](https://docs.qualcomm.com/doc/80-70029-50/topic/gstreamer-application-use-cases.html).
- Follow the [Prerequisites](https://docs.qualcomm.com/doc/80-70029-50/topic/download-model-and-label-files.html) to download the artifacts such as models, labels, and input files required to run the GStreamer command-line use cases.

Important

The AI procedures in this guide are compatible with Qualcomm AI Runtime SDK v2.43 and LiteRT (or TFLite) v2.16.1. Ensure that you download the matching SDKs to your host computer before starting AI/ML development.

- [LiteRT use cases](https://docs.qualcomm.com/doc/80-70029-50/topic/tensorflow-lite-use-cases.html)
LiteRT is a set of tools that allows on-device machine learning. You can run your models on mobile, embedded, and edge devices. LiteRT use cases allow you to run use cases for image classification, object detection, image segmentation, and pose estimation.
- [Qualcomm Neural Processing SDK use cases](https://docs.qualcomm.com/doc/80-70029-50/topic/qualcomm-neural-processing-sdk-use-cases.html)
Qualcomm Neural Processing SDK (formerly known as Qualcomm Snapdragon Neural Processing Engine (SNPE)) is used to run deep neural networks for inference. The use cases describe the image classification, object detection, and image segmentation scenarios using different ML models.
- [ONNX use cases](https://docs.qualcomm.com/doc/80-70029-50/topic/onnx-use-cases.html)
ONNX supports integration of trained machine‑learning models into SDK workflows that require portable, framework‑agnostic inference. It is commonly used in media and analytics pipelines where inference is performed as part of video or audio processing. This approach enables deployment across heterogeneous hardware using a standardized model format.
- [Custom Gstreamer pipeline use cases](https://docs.qualcomm.com/doc/80-70029-50/topic/custom-gstreamer-pipeline-use-cases.html)
Custom Gstreamer pipeline helps you design and implement tailored multimedia processing work flows using the GStreamer framework. These pipelines give you full control over media processing, analysis, and delivery. You can connect specific GStreamer elements and leverage their properties to fine-tune performance, latency, and output formats.

Last Published: Apr 02, 2026

[Previous Topic
GStreamer command-line use cases](https://docs.qualcomm.com/bundle/publicresource/80-70029-50/topics/gstreamer-application-use-cases.md) [Next Topic
LiteRT use cases](https://docs.qualcomm.com/bundle/publicresource/80-70029-50/topics/tensorflow-lite-use-cases.md)