Skip to content

Rakis

Run distributed AI inference from the browser

Visit official site

What Rakis is

The network is described as a decentralized inference system that runs fully in the browser, sharing requests and results between nodes over a blockchain.

What you can do with it

  • Run large language model inference inside a browser
  • Tune the security framework for different scenarios

What to watch out for

  • Nodes need WebGPU-compatible platforms to run models
  • Agreement depends on embeddings and hashes

Pros & cons

✓ What we like

  • No server is needed for distributed execution
  • The project is open source with documentation and community support

! What to watch out for

  • Nodes need WebGPU-compatible platforms
  • Consensus relies on the embedding-based mechanism

FAQ

Where does inference run?

It runs in the browser on WebGPU-compatible platforms.

How are results agreed on?

The material describes embedding consensus with hashes and a commit-reveal system.

Last reviewed: 2026-11-15

More AI coding tools tools

View all →

How we review