Open-source alternatives to AI On The Edge Device

OpenMQTTGateway is a multi-protocol IoT gateway and hardware interface designed for ESP32 and ESP8266 devices. It serves as a bridge that converts Bluetooth Low Energy, Radio Frequency, and Infrared signals into standardized MQTT messages for use in home automation systems. The project provides specific bridging capabilities for 433MHz, 315MHz, and 868MHz radio signals, as well as LoRa and Infrared commands. It includes a Bluetooth presence detector that monitors device advertisements to track the proximity and location of people or objects. The system handles IoT sensor aggregation through

blinker-js is an IoT device communication framework and Node.js library designed to manage bidirectional data exchange between embedded hardware and remote servers or local networks. It functions as an MQTT hardware integration tool that links single-board computers to IoT clouds for remote management and control. The project provides a virtual device widget manager to synchronize hardware state with cloud-based user interface components. This system allows for the mapping of remote UI identifiers to internal software logic, enabling remote IoT control through virtual buttons and input fields

Espectre is an edge machine learning framework and motion detection platform that uses Wi-Fi Channel State Information to identify human presence and movement. It functions as a sensing toolkit for ESP32 microcontrollers, enabling the detection of motion through walls without the use of cameras or wearables. The project distinguishes itself by executing compact neural network classifiers and mathematical detection algorithms directly on the microcontroller. It utilizes a MicroPython runtime to allow for the prototyping and deployment of sensing logic and wireless signal processing algorithms

ExecuTorch is a lightweight C++ runtime for deploying PyTorch models on mobile, embedded, and edge hardware. It provides an ahead-of-time compilation pipeline that exports, quantizes, and lowers model graphs into compact serialized programs, then executes them through a minimal runtime with hardware acceleration and on-device large language model inference capabilities. The project distinguishes itself through a hardware accelerator delegate system that partitions model subgraphs and offloads computation to specialized backends including NPUs, GPUs, and DSPs from Apple, Arm, Intel, MediaTek,

Johnny-Five is a JavaScript robotics framework and microcontroller hardware interface designed for programming robots and IoT devices. It provides a standardized library for managing pins, motors, and displays across various microcontroller platforms, allowing developers to control sensors and actuators using a consistent JavaScript API. The framework is distinguished by its use of a plugin-based hardware abstraction system, which enables communication with diverse hardware platforms and protocols. It supports real-time hardware manipulation and debugging through a read-eval-print loop that a

RocketMQ is a cloud-native distributed messaging platform and streaming engine. It functions as a distributed transactional queue that ensures atomicity between local transactions and message delivery, and serves as an MQTT IoT message broker to bridge lightweight device traffic into high-performance data streams. The system is distinguished by a Kubernetes-native architecture that decouples compute from storage to allow independent scaling of traffic and data retention. It utilizes a tiered storage model to offload older data to remote storage and employs quorum-based replication and automat

Camera.ui is a self-hosted network video recorder that provides a centralized dashboard for managing live surveillance feeds, continuous video recording, and automated security monitoring. It functions as an integrated platform that combines video stream management with an artificial intelligence engine to identify objects and events directly on local hardware. The system distinguishes itself through its event-driven architecture, which utilizes a messaging broker to decouple hardware state updates from automation logic. This allows for deep integration with smart home ecosystems, enabling us

Valetudo is a custom firmware project for robot vacuums that enables local-only control, removing dependencies on cloud servers and protecting user privacy. It replaces proprietary vendor binaries with open source software to ensure that data, including floor plans and images, is not uploaded to external clouds. The project distinguishes itself by providing a rooted firmware installation process that prevents forced vendor updates. It implements a standardized control interface across different hardware brands and utilizes an MQTT-based message bus to facilitate integration with open source a

Home Assistant is a local home automation platform and server that acts as an IoT device orchestrator. It integrates diverse smart home hardware by wrapping third-party APIs into a standardized logic layer and stores all system state and historical statistics on local hardware to eliminate cloud dependencies. The system functions as a Matter IoT controller and an MQTT home automation bridge, allowing for local interoperability between different manufacturers. It features a state-based entity model and an internal event bus that decouple physical device logic from system automation. The platf

MicroPython is a lean implementation of Python 3 optimized to run on microcontrollers and other resource-constrained systems. It serves as a cross-platform embedded runtime and hardware abstraction layer, providing a firmware framework that maps high-level software commands to specific microcontroller registers across diverse processor architectures. The project functions as an embedded language interpreter that enables rapid prototyping on hardware through an interactive read-eval-print loop. It supports a wide range of target environments, including ARM, ESP32, STM32, RISC-V, and WebAssembl

LiteRT is a runtime and API for executing machine learning and generative AI models on mobile, desktop, and IoT hardware. It consists of an inference engine and a specialized environment for running quantized large language and diffusion models locally on edge hardware. The system includes an ahead-of-time model compiler that translates models into hardware-specific bytecode to reduce startup latency and memory overhead. It provides a unified interface for Neural Processing Units with automatic fallback routing to CPUs or GPUs when specific subgraph support is unavailable. An edge model conve

TinyGo is a specialized compiler and development toolkit designed to bring the Go programming language to resource-constrained microcontrollers and WebAssembly environments. It provides a bare-metal runtime environment that enables high-level code execution without the need for a traditional operating system, utilizing an LLVM-based backend to generate efficient machine instructions. The project distinguishes itself through aggressive optimization techniques tailored for small hardware, including a static memory allocation strategy and whole-program dead code elimination that significantly re

libhv is a high-performance C/C++ network library and event-driven I/O framework used to build TCP, UDP, SSL, HTTP, WebSocket, and MQTT clients and servers. It provides a non-blocking event loop for managing network sockets, timers, and system signals across multiple threads. The project is distinguished by its integrated support for specialized network roles, including a full HTTP web server with RESTful routing and middleware, an MQTT messaging client for IoT communication, and the ability to implement SOCKS5 and HTTP proxies. It also features a reliable UDP implementation to ensure ordered

This project provides custom firmware and a suite of management tools for wireless thermometers and hygrometers. It replaces manufacturer software to improve measurement accuracy, extend battery life, and unlock configuration options not available in stock firmware. The system differentiates itself through a comprehensive set of administration utilities, including a device configurator for managing Bluetooth advertising and transmission power, and an over-the-air flasher for wireless software updates. It also features a climate data visualizer to retrieve and graph historical temperature, hum

Gobot is a robotics framework for the Go programming language designed for developing robotics, drones, and IoT applications. It provides a hardware abstraction layer with standardized drivers to interact with GPIO, I2C, SPI, and PWM interfaces across various single-board computers and microcontrollers. The framework functions as an IoT device orchestrator and BLE device manager, enabling the coordination of multiple sensors, actuators, and Bluetooth Low Energy peripherals. It includes specialized interfaces for drone control, allowing for the management of flight maneuvers and video streams

This repository is a comprehensive collection of reference implementations and sample libraries for the Universal Windows Platform. It provides practical examples of how to use Windows Runtime APIs to build cross-device applications, including detailed guidance on XAML-based declarative user interfaces and DirectX-integrated rendering. The project distinguishes itself by providing a wide array of hardware integration suites, covering low-level communication with USB, Serial, I2C, SPI, and GPIO peripherals. It includes specialized implementations for mixed reality holographic rendering, advanc

Mongoose is an embedded networking library providing TCP/IP stacks, web server hosting, and IoT device connectivity. It enables microcontrollers and desktop systems to implement network protocols including HTTP, WebSocket, and MQTT. The project provides a bare-metal TCP/IP stack for environments without a native operating system and includes built-in TLS and ECC encryption to secure network traffic. It features a binary-embedded file system to serve web content directly from the executable and supports over-the-air firmware updates for managing device fleets. Additional capabilities cover as

This project is a Microsoft Teams Linux client that functions as a native desktop wrapper for the web-based communication service. It uses an Electron-based shell to provide system integration and window management for the application on Linux platforms. The client distinguishes itself through a multi-account session manager that uses session partitioning and isolated data directories to allow concurrent access to multiple profiles and tenants. It further integrates with home automation via an MQTT bridge, publishing presence and call status to a message broker and receiving remote commands t

This project is an embedded operating system for creating decentralized peer-to-peer communication networks using long-range radio hardware. It provides an off-grid messaging protocol for transmitting text and GPS coordinates without relying on cellular or internet infrastructure. The firmware implements a secure communication framework using shared keys and high-grade encryption to protect data transmitted over the air. It includes a LoRaWAN hardware interface to manage radio parameters and an MQTT radio gateway to bridge local mesh traffic to external networks. The system covers a broad ra

AISystem is a comprehensive AI full-stack infrastructure project covering the entire pipeline from AI chip architecture to high-level training frameworks. It encompasses the development of AI compiler frameworks, inference engines, and distributed training orchestrators designed to coordinate workloads across a heterogeneous compute stack of CPUs, GPUs, and NPUs. The project focuses on the deep integration of software and hardware, employing software-hardware co-design to align tensor layouts with physical memory structures. It provides specialized capabilities for accelerating Transformer mo

ZSWatch is a real-time operating system for wearable devices featuring a touch display and Bluetooth Low Energy connectivity. It provides a comprehensive software suite that includes a Bluetooth LE communication stack, an embedded UI framework for round displays, and a complete firmware toolchain for compiling, flashing, and simulating wearable hardware. The project is distinguished by its specialized wearable capabilities, such as a graphical framework for designing reactive screens and watchfaces via a visual editor. It implements a publish-subscribe messaging system to decouple hardware se

Vector is a high-performance observability data pipeline designed to collect, transform, and route logs, metrics, and traces across distributed infrastructure. It functions as a modular engine that decouples data ingestion from processing and transmission, utilizing a component-based architecture to connect diverse sources to multiple destinations. The project distinguishes itself through a focus on reliability and flow control. It implements backpressure-aware data movement to prevent data loss during traffic spikes and utilizes disk-backed event buffering to ensure durability during network

ESPHome is a framework for creating and managing custom firmware for microcontrollers, specifically targeting ESP32 and ESP8266 architectures. It replaces the need for writing complex embedded C++ code by allowing users to define hardware behavior, pin configurations, and automation logic through simple, declarative text files. The system automatically compiles these configurations into optimized binary images, providing a streamlined path from design to deployment. The project distinguishes itself through a modular, component-based architecture that emphasizes local-first control, ensuring t

Deep Java Library is a Java deep learning framework and JVM model inference engine. It provides a high-level API for building and deploying deep learning models within the Java ecosystem, acting as a cross-platform runtime for executing models across CPUs, GPUs, and mobile devices. The library is engine-agnostic, allowing users to switch between different deep learning engines such as PyTorch, TensorFlow, and MXNet while maintaining a single unified API. This enables the deployment of the same model across different backends without changing the application code. The framework supports the f

Embassy is an asynchronous framework for microcontrollers that provides a runtime for cooperative multitasking. It enables the execution of multiple concurrent tasks to maximize energy efficiency and CPU usage, utilizing an async executor that puts the processor to sleep when idle. The framework includes a comprehensive set of embedded components, featuring a hardware abstraction layer for peripherals like GPIO, SPI, I2C, and UART, alongside a TCP/IP network stack and a USB device stack. It also provides a Bluetooth Low Energy host stack for wireless communication and a bootloader framework t

This project is network firmware for the ESP32 that functions as a wireless repeater, NAT router, or wireless bridge to extend WiFi coverage. It can operate as a WiFi range extender, a layer-2 bridge for shared subnets, or a NAT router that creates a private secondary subnet to connect clients to an upstream network. A key differentiator is its ability to act as an enterprise WiFi to PSK converter, translating PEAP-authenticated networks into WPA2-PSK networks for devices that do not support enterprise authentication. Additionally, it includes a network traffic mirroring tool that streams int

This repository is a collection of sample applications and reference projects for integrating on-device machine learning APIs into mobile applications. It provides implementation examples for both Android and iOS, demonstrating how to embed computer vision and natural language processing models. The project covers the implementation of mobile computer vision, including text recognition, face detection, and barcode scanning. It includes workflows for real-time image processing and the integration of pre-trained models that process data locally on the device.