Get Started

Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI

LaunchLLM InfrastructureBrowsers & Extensions

First reported · 53d ago · 1 source

Crest summary — written by AI from the 1 source below; every figure is checked against them · model deepseek-chat

Summary

Hugging Face released @huggingface/kernels, a library of more than 200 WebGPU kernels for running AI models locally in the browser.

Key facts

  • Hugging Face introduced @huggingface/kernels on 2026-09-01.
  • The library contains over 200 WebGPU kernels.
  • The kernels are intended for local AI execution.

Why it matters

Developers can run AI models locally through WebGPU instead of relying on server-side inference, which may reduce dependence on remote compute for browser-based applications.

Companies

Technologies

  • @huggingface/kernelslibrary
  • WebGPUstandard

Original sources (1)

The report this event was built from. Crest of Tech keeps no article bodies: the full text is at each source.

  1. Hugging FaceOfficial · Primary
    Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI
    Read at Hugging Face
Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI — Crest of Tech