A curated list of hardware, software, frameworks, and resources for developing Artificial Intelligence at the edge. It serves engineers and researchers by providing a centralized overview of components for embedded AI, from specialized chips to optimized libraries, facilitating faster exploration and adoption of edge AI technologies.

How It Works

The project functions as a categorized directory, detailing specialized AI hardware (ASICs, SoCs, MCUs with accelerators) and a wide array of software solutions. It highlights inference engines, development platforms, and compiler toolchains designed for resource-constrained environments, enabling on-device ML deployment. This curated approach offers a broad landscape view of the edge AI ecosystem.

Quick Start & Requirements

This repository is a curated list of resources, not a deployable project. Users must select and integrate individual components. Requirements vary significantly based on chosen hardware (e.g., NVIDIA Jetson, Google Coral, Kendryte K210) and software (e.g., TensorFlow Lite, Edge Impulse). Links to official documentation for each component are typically provided within the README.

Highlighted Details

• Extensive coverage of specialized AI hardware accelerators (e.g., Google Edge TPU, Intel Movidius, Kendryte K210/K510, ARM Ethos-U55).
• Broad spectrum of software solutions, including optimized inference engines (TensorFlow Lite, emlearn), development platforms (Edge Impulse), and compiler toolchains (nncase, uTVM).
• Strong emphasis on microcontroller-based AI (TinyML) alongside more powerful embedded systems.

Maintenance & Community

• The repository includes a "Contributing" section with guidelines for adding new resources, encouraging unbiased descriptions and individual commits.
• It appears to be community-driven through contributions.

Licensing & Compatibility

• The work is dedicated to the public domain under CC0 ("To the extent possible under law, Xabi Crespo has waived all copyright and related or neighboring rights to this work.").
• This license permits unrestricted use, modification, and distribution, including for commercial purposes, without copyleft or attribution requirements.

Limitations & Caveats

As a curated list, this repository lacks a unified framework or direct implementation. Users must independently evaluate and integrate components. The rapidly evolving edge AI field means some listed resources may become outdated, and the README does not explicitly track versioning or deprecation status for all items.