Frequently Asked Questions (FAQ)
General Questions
What is Horizons OmniChat?
Horizons OmniChat is an open-source chatbot platform that brings enterprise-grade LLM capabilities to your infrastructure, with flexible deployment options including local, hybrid, and AWS modes.
What makes Horizons different from other chatbot platforms?
- Complete privacy control with on-premises deployment options
- Multiple deployment modes (Local/Hybrid/AWS)
- Enterprise-grade security and compliance features
- Integration with both ollama and cloud LLM models (Bedrock)
- Open-source with enterprise support options
Which deployment mode should I choose?
| Mode | Best For | Requirements | Key Benefits |
|---|---|---|---|
| Local | Development, Testing, Privacy | 8GB RAM, Docker | Complete Control |
| Hybrid | Production, Cost-Effective | AWS Account, 8GB RAM | Flexibility |
| AWS | Enterprise, Scalability | AWS Account | Full Cloud Benefits |
Technical Questions
What are the system requirements?
See our detailed System Requirements guide, but in general:
- Minimum 8GB RAM
- 4+ CPU cores
- 20GB storage
- Docker/Podman
- (Optional) NVIDIA GPU
Which models are supported?
Local Models (via Ollama)
- Llama 2
- Mistral
- TinyLlama
- Deepseek
- Qwen
- ALIA/Salamandra
- Custom models
Cloud Models (via AWS Bedrock)
- Claude (Anthropic)
- Titan (Amazon)
- Nova (Amazon)
- Jurassic (AI21)
- Command (Cohere)
How do I update to the latest version?
# Local/Hybrid Mode
git pull origin main
make local-down # or hybrid-down
docker compose pull
make local-up # or hybrid-up
# AWS Mode
git pull origin main
make aws-plan
make aws-apply
Deployment Questions
Can I deploy Horizons in my own datacenter?
Yes! The Local and Hybrid modes are specifically designed for on-premises deployment.
How do I scale Horizons?
Local/Hybrid Mode
- Vertical scaling (more CPU/RAM)
- GPU acceleration
- Connection pooling
AWS Mode
- Auto-scaling groups
- Load balancing
- Multi-AZ deployment (optional)
- Elastic scaling
How do I backup my data?
See our detailed Backup Guide, but in general:
# Local/Hybrid Mode
docker exec open-webui-db pg_dump -U $POSTGRES_USER $POSTGRES_DB > backup.sql
# AWS Mode
aws rds create-db-snapshot --db-instance-identifier horizons-persistence-db
Security Questions
How is data protected?
- End-to-end encryption
- Data encryption at rest
- TLS 1.3 for all communications (AWS mode)
- Role-based access control
- Audit logging (basic)
- See our Security Overview
Does Horizons comply with GDPR/HIPAA/SOC2?
- GDPR compliance features included
- ENS compliance possible with Enterprise edition
- HIPAA compliance possible with Enterprise edition
- SOC 2 controls available in Enterprise edition
- See our Compliance Guide
How are updates and security patches handled?
- Regular security updates
- Automated vulnerability scanning
- Dependency updates
- Security advisories
- See our Security Best Practices
Enterprise Questions
What enterprise features are available?
- Advanced security controls
- High availability deployment
- Premium support
- Custom development
- Compliance assistance
- Enhanced security and compliance
- Customised presentation layer
- Fine-grained customisation
- Knowledge Database powered by specific user/business data
- See our Enterprise Guide
How do I get enterprise support?
Contact our Enterprise Support Team for:
- Dedicated support
- SLA guarantees
- Custom features
- Training
- Consulting
Can I customize Horizons for my organization?
Yes! Options include:
- Custom model integration
- UI customization
- Authentication integration
- Custom workflows
- See our Development Guide
Troubleshooting
Common Issues
Model Loading Issues
# Check Ollama status
docker logs ollama
docker exec ollama ollama list
Performance Issues
# Check resource usage
docker stats
nvidia-smi # if using GPU
Connection Issues
# Verify services
docker compose ps
curl http://localhost:3002/health
See our Troubleshooting Guide for more details.
Getting Help
Where can I find documentation?
How do I report issues?
- Check existing GitHub Issues
- Create a new issue with:
- Deployment mode
- Error messages
- Steps to reproduce
- System information
How do I contribute?
See our Contributing Guide for:
- Development setup
- Coding standards
- Pull request process
- Community guidelines
Where can I get community support?
Horizons OmniChat by evereven