Introduction

Introduction#

Welcome to Mobilint’s SDK qb documentation page!

This guide aims to help developers, engineers, and researchers quickly set up Mobilint’s software development kit, or SDK qb, and run their first AI inference on supported hardware.

Whether you are building a prototype on a desktop or deploying on an embedded AI system in the field, SDK qb provides the essential software building blocks to harness the full compute potential of Mobilint’s hardware.

What is SDK qb?#

SDK qb is Mobilint’s software development kit designed to enable efficient and optimized utilization of Mobilint’s Neural Processing Unit (NPU).

SDK qb is composed of below three proprietary modules that bring out the full performance of Mobilint’s NPU products and enable seamless integration in deployment processes.

  • Driver - the bridge between the software and the hardware

  • qb Runtime - the module provides convenient & fast inference using internal optimized logic

  • qb Compiler - the module translates deep learning model into Mobilint’s eXeCUtable, MXQ

These modules are available in flexible combinations depending on the form factor and core NPU involved.

Technical Overview#

The following diagrams are an overview of how SDK qb’s components work within the target application and device.

Model Build

Runtime Execution

To enable easier system integration and maximize user convenience, SDK qb supports both Linux (Ubuntu) and Windows OS. It is compatible with major deep learning model frameworks, such as TensorFlow, TFLite, ONNX, KERAS, and PyTorch. Additionally, for developers’ ease, the runtime library provides accessible interfaces in both C++ and Python.

Mobilint Model Zoo#

In addition, Mobilint provides a growing Model Zoo of open-source models precompiled with Mobilint’s qubee compiler.

With over 400 models across computer vision, natural language processing, and multimodal AI, these .mxq (Mobilint ExeCUtable) files are optimized for Mobilint’s runtime and can be deployed immediately on supported NPU hardware.

New models are added regularly to support evolving AI workloads and customer requests.

Get Started#

Click here to get started, from installing the product to maximizing its performance with advanced functions.

Have questions or need custom deployment support? Contact us at tech-support@mobilint.com.

Contents