Helix Documentation
Getting Started
Helix Quick Start

Helix Quick Start

Helix is designed for private deployment on your own infrastructure. This guide will help you get started quickly using the Helix installer.

Prerequisites

Before you begin, ensure you have:

  • A Linux machine (Ubuntu 20.04+ recommended)
  • Docker installed
  • Root or sudo access for installation
  • (Optional) An NVIDIA GPU for running local models

Quick Start

1. Download and Run the Installer 

curl -sL -O https://get.helixml.tech/install.sh
chmod +x install.sh
sudo ./install.sh

The installer will guide you through configuration options and set up all required services.

2. Access Helix

Once installation completes, Helix will be available at:

  • Frontend: http://localhost:8080
  • API: http://localhost:8080/api/v1

3. Install the Helix CLI (Optional)

Install the CLI to interact with Helix from the command line:

sudo ./install.sh --cli

Connect the CLI to your Helix instance:

export HELIX_URL=http://localhost:8080
export HELIX_API_KEY=your-api-key

Get your API key from Account → API Keys in the web UI.

4. Connect a GPU Runner (Optional)

If you have an NVIDIA GPU and want to run local models, attach a runner:

  1. Log into Helix at http://localhost:8080
  2. Go to Account → Runners to find your runner token
  3. Start the runner with your token:
docker run --gpus all \
  -e RUNNER_TOKEN=your-token-here \
  -e API_HOST=http://your-helix-server:8080 \
  registry.helixml.tech/helix/runner:latest

Alternatively, use external LLM providers (OpenAI, Anthropic, Together) without a local GPU. Configure these in Account → AI Providers.

What You Can Do with Helix

Chat with AI Models

Once deployed, you can interact with AI models through the web interface or API:

  • Text conversations with models like Llama, Qwen, or external providers
  • RAG (Retrieval Augmented Generation) - Upload documents and chat with your data
  • Fine-tuning - Train models on your own text data

Build AI-Powered Apps

Create custom AI applications using:

  • Helix Agents - Define tools, knowledge bases, and system prompts
  • API Integration - Use the OpenAI-compatible API in your applications

Coding Agents

Run fleets of background coding agents with real GPU-accelerated desktops:

  • Desktop streaming - Watch AI agents work in real-time via browser
  • Spec-first development - Agents write design docs before code
  • Kanban orchestration - Manage dozens of agents from one view

Next Steps

Getting Help

If you encounter issues:

Last updated on
Frequently asked questions