Self-Hosted Privacy

For users with strict privacy requirements, Chatelope can be run entirely self-hosted with local AI models, ensuring your data never leaves your infrastructure.

Why Self-Host?

Data Sovereignty

• Conversations stay on your hardware
• No data sent to cloud providers
• Full control over storage and backup

Compliance

• Meet regulatory requirements (GDPR, HIPAA)
• Avoid third-party data processing
• Maintain audit trails

Security

• No API keys to protect
• No network exposure
• Air-gapped deployment possible

Architecture Options

Fully Local

Everything runs on your machine:

Your Computer
├── Chatelope (web app)
├── SQLite database
└── Ollama (local AI)

Private Server

Run on your own server:

Your Server
├── Chatelope (Docker)
├── PostgreSQL
└── Ollama / LocalAI

Enterprise

Deploy in your infrastructure:

Your Network
├── Chatelope cluster
├── Central database
├── AI inference server
└── Backup storage

Setup Guide

Local Desktop Setup

1. Install Ollama

# macOS
brew install ollama

# Linux
curl -fsSL https://ollama.com/install.sh | sh

# Windows
# Download from ollama.com

2. Pull Models

ollama pull llama3
ollama pull mistral

3. Run Chatelope

# Clone repository
git clone https://github.com/chatelope/chatelope
cd chatelope/web

# Install dependencies
npm install

# Configure for local
cp .env.example .env.local
# Edit .env.local - set Ollama as default provider

# Start
npm run dev

4. Configure

1. Open http://localhost:3000
2. Go to Settings > AI Settings
3. Select Ollama as provider
4. Verify connection to local server

Docker Deployment

docker-compose.yml

version: '3.8'

services:
  chatelope:
    build: ./web
    ports:
      - "3000:3000"
    volumes:
      - ./data:/app/data
    environment:
      - DATABASE_URL=file:/app/data/chatelope.db
      - DEFAULT_PROVIDER=ollama
      - OLLAMA_URL=http://ollama:11434

  ollama:
    image: ollama/ollama
    ports:
      - "11434:11434"
    volumes:
      - ollama_data:/root/.ollama

volumes:
  ollama_data:

Start Services

docker-compose up -d
docker exec ollama ollama pull llama3

Privacy Features

No Telemetry

Chatelope collects no analytics:

• No usage tracking
• No error reporting
• No feature flags

Encrypted Storage

Enable database encryption:

1. Go to Settings > Data
2. Enable Encryption
3. Set a strong password

Network Isolation

For maximum privacy:

1. Block outgoing network connections
2. Use local AI models only
3. Disable update checks

Compliance Checklist

GDPR

• Data stored locally
• No third-party processors
• Export functionality available
• Delete functionality available

HIPAA

• Encryption at rest
• Access logging
• Backup procedures
• Security assessment

SOC 2

• Access controls
• Audit logging
• Change management
• Incident response

Backup Strategy

Local Backups

# Backup database
cp data/chatelope.db backups/chatelope-$(date +%Y%m%d).db

# Backup attachments
tar -czf backups/attachments-$(date +%Y%m%d).tar.gz data/attachments/

Encrypted Offsite

For disaster recovery without cloud exposure:

1. Encrypt backups locally
2. Transfer to secure offsite storage
3. Test restoration regularly

Air-Gapped Deployment

For the highest security:

Initial Setup

1. Download Chatelope on connected machine
2. Download Ollama models
3. Transfer via secure media
4. Install on air-gapped machine

Updates

1. Download updates on connected machine
2. Security review of changes
3. Transfer via secure media
4. Apply updates offline

Tips

• Use hardware with local AI capability (GPU)
• Regular security updates for all components
• Document your deployment for audit
• Test backup restoration regularly
• Consider hardware security modules (HSM) for keys