Skip to main content
Page is under construction.
Feedback Welcome!
Complete: needs review

Gateway Modes

You can run a gateway
  • Off-chain -> dev or local mode
  • On-chain -> production mode connected to the blockchain-based Livepeer network.

Gateway Capabilities

You can run a Gateway for:
  • Video Only -> traditional transcoding services
  • AI Only -> AI inference services
  • Dual: AI & Video -> both video transcoding and AI inference services

Quick OS Table

OSInstall MethodGatewayOrchestrator / AI WorkerNotes
macOS (Intel / Apple)Build from source (Go)✅ Yes❌ NoBest for local dev & routing. Docker often fails on macOS.
Windows (Native)Not supported❌ No❌ NoNo native Windows binaries intended for production.
Windows (WSL2 + Docker)Docker (Linux via WSL2)✅ Yes⚠️ Limited / fragileEffectively Linux-in-WSL. GPU needs NVIDIA + CUDA for WSL.
Linux (Ubuntu recommended)Docker or source build✅ Yes✅ Yes (NVIDIA GPU)Only OS suitable for production Orchestrators & AI workers.

Gateway Set Up Requirements

Technical Knowledge

Dual AI & Video A basic understanding of system administration and CLI tooling… or know how to ask AI for help :)

Hardware Requirements

Dual AI & Video Gateways do not require GPU resources, but must be able to handle high network throughput. This is necessary to route requests between applications and Orchestrators.
A typical setup includes
  • 4–8 CPU cores
  • 16–32 GB RAM
  • High-speed NVMe (optional, recommended)
  • Stable multi-region networking
  • Linux or containerised deployment environment

Network Requirements

Dual AI & Video Your Gateway must be reachable and responsive.
  • Public HTTPs endpoint
  • Low-latency connectivity to Orchestrators
  • Ability to handle high request throughput
Recommended:
  • Multi-region deployment or failover
  • Static IP or domain name (required for production)

OS & Software Requirements

Gateways install the Go-Livepeer Gateway Software.Installation requires the following OS with root or sudo access:
Video Only Gateway:   
  • Linux, Windows or Docker containerised deployment environment
AI or Dual Gateway:
  • Linux OS (Windows and macOS support coming soon)

Installation is covered in detail in theInstallation Guide

Additional Requirements

Dual AI & Video You can run a gateway
  • Off-chain -> dev or local mode
  • On-chain -> production mode connected to the blockchain-based Livepeer network.
Additional requirements apply when running a Gateway in off-chain mode:You will need to run an Orchestrator Node locally to test Gateway functionality.
This requires that you
Have a GPU available (locally or cloud) with root or sudo access
Setup the GPU as an Orchestrator per the Orchestrator Setup Guide
Off-chain mode is ideal for:
  • Local development and testing
  • Learning the gateway functionality
  • Prototyping without financial commitments
  • Running private transcoding infrastructure
Last modified on January 13, 2026