Skip to main content
Version: nightly


Explore these foundational guides that will provide essential information and instructions on core concepts and techniques used across Neural Magic. As you grasp the underlying principles and best practices, you are encouraged to dig in on the domain-specific use cases. Product-specific guides are also provided.

📄️ Sparsification

A comprehensive overview of sparsification techniques used to create smaller, faster, and more energy-efficient neural networks while maintaining accuracy.

🗃️ DeepSparse Features

6 items

🗃️ Deployment Options

4 items


Overview of ONNX, its role in DeepSparse for CPU inference, and guidance on exporting models to the ONNX format.