webmachinelearning/webnn

🧠 Web Neural Network API

What it solves

This project defines a standardized API for neural networks on the web, allowing developers to run machine learning models efficiently across different browsers and hardware. It aims to provide a consistent way for web applications to access hardware acceleration for AI tasks.

How it works

It is a developed as a W3C community group specification. The repository contains the source files (written in Bikeshed syntax) that are converted into a formal HTML specification document. This document defines the rules and interfaces that browser vendors must implement to enable high-performance machine learning in the browser.

Who it’s for

Web developers and browser engineers who want to implement or use a standardized interface for running neural networks in web applications.

Highlights

  • Standardized Web Neural Network API specification.
  • Developed by the Web Machine Learning Working Group.
  • Uses Bikeshed for specification generation.
  • Aims for cross-browser and hardware acceleration compatibility.

Related

  • Project
    gpuweb/gpuweb

    The official W3C repository for the WebGPU API and WebGPU Shading Language (WGSL) specifications, enabling high-performance GPU acceleration for web applications.

  • Project
    webmachinelearning/prompt-api

    A proposed web platform API that gives developers a direct, standardized way to access built-in browser or OS language models for local, private, and cost-effective AI integration.

  • Project
    agentrhq/webcmd

    A self-learning browser infrastructure for AI agents that converts repetitive web navigation into deterministic CLI commands to reduce token costs and increase reliability.

  • Project
    huggingface/transformers

    A model-definition framework providing a unified API to access, train, and deploy state-of-the-art pretrained models across text, vision, audio, and multimodal domains.

  • Project
    KhronosGroup/OpenCL-Headers

    The official C language headers for the OpenCL API, providing a standardized interface for developing high-performance applications on heterogeneous computing platforms.