Deploying large AI models to tiny embedded devices like Arduino and Raspberry Pi requires aggressive model slimming through quantization, pruning, and distillation, careful selection of runtimes such as TensorFlow Lite, and addressing power, latency, and debugging challenges to achieve real‑time inference.
This article explains the challenges of running large language models on mobile devices, reviews recent industry efforts, and provides a step‑by‑step guide—including code snippets—for integrating a distilled GPT‑2 model with Sohu's Hybrid AI Engine using TensorFlow Lite and Keras‑NLP for on‑device inference.
This article explains three traditional approaches to on‑device AI for mobile apps, compares their drawbacks, and introduces Sohu News' Hybrid AI Engine that unifies TensorFlow Lite and system‑level AI (HiAI) with concise code examples and future open‑source plans.
The article details how edge intelligence is applied to the Vivo official app to improve product recommendation on the smart‑hardware floor by abstracting the problem, designing feature engineering pipelines, training TensorFlow models, converting them to TFLite, and deploying inference on mobile devices, while also covering monitoring and performance considerations.
This article details how Vivo’s official app implements edge intelligence to dynamically rank and recommend hardware products on its homepage, covering problem abstraction, data collection, feature engineering, model design, TensorFlow‑Lite conversion, on‑device inference, and monitoring for a personalized user experience.
Google I/O highlighted the rise of on‑device AI, showing how new neural processors, Edge TPU, and tools like the Edge AI SDK and TensorFlow Lite enable developers to run large language models locally, reducing latency, cost, and privacy concerns while integrating with cloud resources.
This guide walks you through the end‑to‑end process of building, training, converting, and deploying a TensorFlow Lite text‑classification model on Android, covering data preparation, model selection, performance trade‑offs, and integration using the TFLite Task Library.
The article outlines how Sohu’s Hybrid AI Engine enables on‑device deployment of a distilled GPT‑2 model by converting it to TensorFlow Lite, detailing the setup, customization with Keras, inference workflow, and core SDK calls, and argues that this approach offers fast, private, and cost‑effective AI for mobile devices despite typical LLM constraints.
This article explains how front‑end developers can run AI locally on web, Android, iOS and Raspberry Pi devices using TensorFlow Lite MoveNet for real‑time pose estimation, walks through setup, model variants, Python code examples, and practical applications such as yoga‑pose classification.
This article explains how to convert a TensorFlow image‑classification model to TensorFlow Lite, apply different quantization techniques, benchmark the resulting models on a Raspberry Pi 4, and compare latency, size, and accuracy to demonstrate the trade‑offs of edge AI deployment.
This article introduces three traditional approaches for deploying machine‑learning models on mobile devices, analyzes their drawbacks, and presents a hybrid AI engine that combines TensorFlow Lite and system‑level HiAI to provide a unified, lightweight, and developer‑friendly on‑device image‑recognition solution, including code examples.
Android Studio 4.1 introduces a suite of enhancements—including a Database Inspector, Material Design component updates, integrated TensorFlow Lite support, improved Apply Changes, native memory profiling, and expanded emulator capabilities—aimed at boosting productivity and code quality for Android developers.
Android Studio 4.1 stable release introduces upgraded Material Design components, a built‑in Database Inspector, direct Android Emulator support, Dagger navigation, TensorFlow Lite model integration, foldable‑device emulator capabilities, and a native memory profiler, enhancing the overall Android development experience.
This article explains how to train and deploy convolutional neural networks directly on iOS and Android devices using Core ML, NNAPI, and TensorFlow Lite, compares performance with desktop TensorFlow, and provides practical code snippets and build‑time tips for mobile AI development.
The article reviews four mobile‑focused machine‑learning frameworks—NCNN, TensorFlow Lite, PyTorch Mobile (Caffe2) and FeatherKit—detailing their size, speed, and resource trade‑offs, and explains Q Music’s edge‑inference pipeline, optimization strategies, and the challenges of performance variability on heterogeneous mobile devices.
This article details how the 58 Used‑Car mobile team built, trained, and optimized TensorFlow‑based object‑detection models for on‑device vehicle and VIN code recognition, covering data preparation, model conversion to TF‑Lite, performance improvements, engineering integration on Android/iOS, and real‑world deployment results.
SmileAR, iQIYI’s self‑developed mobile AR platform powered by TensorFlow Lite, delivers real‑time face, body and gesture recognition across iQIYI’s apps through MobileNet‑based models, quantization‑aware training, multi‑task learning and encrypted SDKs, achieving fast, lightweight, cross‑platform AR experiences for millions of users.
Xianyu leverages a custom TensorFlow Lite framework to power AI‑driven features such as dynamic video‑cover selection, video fingerprinting, and furniture recognition for smart rentals, while its UI2Code tool transforms screenshots into pixel‑perfect production UI code, emphasizing extensibility, security, and online model updates.
This article details the engineering of MobileNet V1 for image classification on mobile terminals, covering its depthwise separable convolution architecture, data collection and preprocessing, model training with transfer learning, TensorFlow Lite conversion, deployment on iOS/Android, and GPU acceleration techniques for faster inference.
The article presents a complete iOS pipeline that trains an SSD‑MobileNet detector with TensorFlow, optimizes and converts it to TensorFlow Lite, runs C++ inference, applies non‑maximum suppression, and finally removes detected watermarks using OpenCV inpainting or pixel‑wise inversion, while discussing practical limitations.
The article provides an engineering‑focused guide to creating a custom TensorFlow Lite operation pipeline, covering its definition, server‑side registration and compilation, client‑side downloading, verification, decryption and dynamic loading, and discusses current limitations and possible extensions such as compression and new tensor types.
This article explains how to implement client‑side human voice recognition on mobile devices using TensorFlow Lite, detailing the mel‑spectrogram feature extraction, algorithmic optimizations such as ARM instruction set and multithreading, model selection with Inception‑v3 CNN, training procedures, and deployment steps.
The paper presents a client‑side speech recognizer that uses a compact TensorFlow Lite Inception‑v3 CNN model combined with an optimized MFCC feature pipeline and ARM‑NEON‑accelerated, multi‑threaded processing, achieving low‑latency, high‑accuracy voice recognition on mobile and embedded devices.
