Quantara — AI Hosting Lab

// 01 — Overview

What Is OpenClaw AI?

OpenClaw is a self-hosted AI gateway that lets you connect to any large language model provider through a single unified interface. Run it on your own server, laptop, or cloud VM — your conversations never leave your infrastructure.

🌐

One Interface, Every Model

Switch between Claude, GPT-4o, Gemini, DeepSeek, Llama, Mistral, and more with a single dropdown. No need for separate tabs or subscriptions — just API keys.

🔒

Full Data Sovereignty

Your prompts, conversations, and files stay on your hardware. Nothing is logged by third-party dashboards. API calls go direct to the provider and back to you.

⚡

Docker-First Architecture

A single docker run command gets you running in under 60 seconds. Auto-updates, persistent storage, and reverse-proxy ready out of the box.

// 02 — Installation

Quick Start Guide

Get OpenClaw running locally or on a remote server in three steps.

1

Install Docker

Docker is the only prerequisite. Install it on any OS.

Terminal — Install Docker

# macOS $ brew install --cask docker # Ubuntu / Debian $ curl -fsSL https://get.docker.com | sh $ sudo usermod -aG docker $USER # Windows: Download Docker Desktop from docker.com

2

Launch OpenClaw

Pull and run the container. Data persists between restarts.

Terminal — Launch OpenClaw AI

$ docker run -d -p 3000:8080 \ -v openclaw-data:/app/backend/data \ --name openclaw --restart always \ ghcr.io/open-webui/open-webui:main # OpenClaw is now running at: http://localhost:3000 # Create your admin account on first visit

3

Connect Your First AI

Go to Settings → Connections and add any provider API key. Models will auto-populate.

OpenClaw UI — Add Your First Provider

# Navigate to: Settings → Connections # Click "Add New Connection" Name: Anthropic Claude Base URL: https://api.anthropic.com/v1 API Key: sk-ant-your-key-here # Models will auto-populate in the model dropdown # Select a model and start chatting!

+

With Local AI (Ollama)

Run completely offline by connecting to a local Ollama instance. Zero cloud dependency.

Terminal — OpenClaw + Ollama (Fully Local)

# Step 1: Install and start Ollama $ curl -fsSL https://ollama.com/install.sh | sh $ ollama pull llama3.2 # Step 2: Launch OpenClaw with Ollama connection $ docker run -d -p 3000:8080 \ --add-host=host.docker.internal:host-gateway \ -e OLLAMA_BASE_URL=http://host.docker.internal:11434 \ -v openclaw-data:/app/backend/data \ --name openclaw --restart always \ ghcr.io/open-webui/open-webui:main # All conversations stay 100% on your machine

// 03 — Connect Any AI Provider

Supported Providers

OpenClaw supports the OpenAI-compatible API format. Any provider that uses this format works out of the box. Here's how to connect each one.

C

Anthropic Claude

Claude Opus, Sonnet, and Haiku. Get your key from console.anthropic.com

Claude Setup

Base URL: https://api.anthropic.com/v1 API Key: sk-ant-api03-xxxxx # Models: claude-opus-4-6, claude-sonnet-4-5, claude-haiku-4-5

O

OpenAI (GPT-4o, o1)

Native support. Get your key from platform.openai.com

OpenAI Setup

Base URL: https://api.openai.com/v1 API Key: sk-proj-xxxxx # Models: gpt-4o, gpt-4-turbo, o1-preview, gpt-4o-mini

G

Google Gemini

Free tier available. Get your key from aistudio.google.com

Gemini Setup

Base URL: https://generativelanguage.googleapis.com/v1beta/openai API Key: AIzaSyXXXXXX # Models: gemini-2.0-flash, gemini-1.5-pro

D

DeepSeek (R1, V3)

Ultra-cheap reasoning models. Key from platform.deepseek.com

DeepSeek Setup

Base URL: https://api.deepseek.com/v1 API Key: sk-xxxxx # Models: deepseek-reasoner, deepseek-chat

G

Groq (Ultra-Fast)

Lightning-fast inference on LPU chips. Free tier at console.groq.com

Groq Setup

Base URL: https://api.groq.com/openai/v1 API Key: gsk_xxxxx # Models: llama-3.3-70b, mixtral-8x7b, gemma2-9b

M

Mistral AI

European AI. Key from console.mistral.ai

Mistral Setup

Base URL: https://api.mistral.ai/v1 # Models: mistral-large, codestral, mistral-small

P

Perplexity (Search AI)

AI with real-time web search. Key from perplexity.ai

Perplexity Setup

Base URL: https://api.perplexity.ai # Models: sonar-pro, sonar, sonar-reasoning-pro

// 04 — Advanced Configuration

Production Hosting

Deploy OpenClaw to a cloud server and access it from anywhere, securely.

☁

Cloud Server Deployment

Host on any VPS (Hetzner, DigitalOcean, Linode, AWS) with a custom domain and SSL.

Server — Nginx Reverse Proxy + SSL

# Install Nginx and Certbot $ sudo apt install nginx certbot python3-certbot-nginx # Create Nginx config: /etc/nginx/sites-available/openclaw server { server_name ai.yourdomain.com; location / { proxy_pass http://localhost:3000; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header Upgrade $http_upgrade; proxy_set_header Connection "upgrade"; } } # Enable site and get SSL certificate $ sudo ln -s /etc/nginx/sites-available/openclaw /etc/nginx/sites-enabled/ $ sudo certbot --nginx -d ai.yourdomain.com $ sudo systemctl reload nginx

🔄

Docker Compose (Multi-Service)

Run OpenClaw with Ollama and auto-updates in a single compose file.

docker-compose.yml

version: "3.8" services: ollama: image: ollama/ollama:latest ports: ["11434:11434"] volumes: [ollama:/root/.ollama] restart: always openclaw: image: ghcr.io/open-webui/open-webui:main ports: ["3000:8080"] environment: - OLLAMA_BASE_URL=http://ollama:11434 volumes: [openclaw-data:/app/backend/data] depends_on: [ollama] restart: always volumes: ollama: openclaw-data: # Launch: docker compose up -d

💡 Pro Tip — Multi-Provider Strategy

Add all your provider keys at once. Use a fast local model (Llama 3.2 7B) for quick questions, Claude Sonnet for complex reasoning, and GPT-4o for image analysis. OpenClaw lets you switch with a single dropdown — every conversation stays in one place.