Overview

This guide walks you through connecting Ollama to Matrix using OpenClaw. You'll learn how to configure authentication, set up the channel, and deploy your AI assistant.

Ollama Features

  • Completely local
  • No API costs
  • Privacy-focused
  • Many model options

Matrix Capabilities

  • Direct messages
  • Group chats
  • Media support
  • Reactions
  • Thread support

Step 1: Configure Ollama

No authentication required for local instance

  1. Install Ollama from ollama.ai
  2. Run 'ollama pull llama3.3'
  3. Start Ollama service
  4. Configure baseUrl in OpenClaw

Step 2: Configure Matrix

  1. Create a Matrix account on your preferred homeserver
  2. Generate an access token
  3. Configure matrix in openclaw.json
  4. Start the gateway
  5. Invite the bot to rooms

Step 3: Combined Configuration

Add both configurations to your openclaw.json:

{
  "agents": {
    "defaults": {
      "model": {
        "primary": "ollama/llama3.3"
      }
    }
  },
  "models": {
    "providers": {
      "ollama": {
  "models": {
    "providers": {
      "ollama": {
        "baseUrl": "http://localhost:11434"
      }
    }
  },
  "channels": {
    "matrix": {
      "homeserver": "https://matrix.org",
      "accessToken": "${MATRIX_ACCESS_TOKEN}",
      "userId": "@mybot:matrix.org"
    }
  }
}
}

Step 4: Start the Gateway

# Start the gateway
openclaw gateway start

# Check status
openclaw status

# View logs
openclaw logs --follow

Access Control

Matrix supports the following access control policies:

DM Policies

Policy Description
allowlist Only senders in allowFrom list are processed
pairing Unknown senders receive a pairing code; admin must approve
open All DMs are processed (requires allowFrom: ["*"])

Group Policies

Policy Description
allowlist Only groups in groupAllowFrom are processed
open All groups are processed

Deploy Options

Choose how to deploy your Ollama + Matrix setup:

Local Deployment

Run on your personal machine with local-only access

View Guide →

VPS Deployment

Always-on deployment on a Linux VPS

View Guide →

Cloud Deployment

Managed container deployment on Fly.io or similar

View Guide →

Docker Deployment

Containerized deployment with Docker

View Guide →

Frequently Asked Questions

How do I set up Ollama with Matrix?
Configure Ollama as your AI provider and enable Matrix as a channel in openclaw.json. The gateway routes Matrix messages to Ollama for processing automatically.
Is Ollama a good choice for Matrix bots?
Ollama works great with Matrix. No API costs and Full data privacy make it well-suited for Privacy-focused teams and Open source communities.