AI Customer Support, Built Right
From AI-powered responses to enterprise security - everything your support team needs in one platform, without the enterprise price tag.
Multi-LLM AI Routing
Choose the best model for each conversation. Chatonn routes across OpenAI GPT-4, Anthropic Claude, Google Gemini, Groq, and Ollama - with automatic fallback and cost routing.
- Bring your own API keys or use AI-managed responses
- Auto cost-routing picks the cheapest capable model
- Per-conversation model assignment
- No vendor lock-in
RAG Knowledge Base
Train your bot on the content that matters: PDFs, docs, FAQ pages, help-center exports, and your whole website. Answers are grounded in your content with visible sources.
- Upload PDFs, docs, and structured FAQ files
- Crawl your website and help center
- Up to 50 knowledge sources on Pro, unlimited on Business
- Citations so customers can verify answers
NL-to-SQL Database Queries
Let customers ask natural-language questions about their data. Chatonn converts them to SQL and returns real-time results - with schema explorer and table access control.
- Connect PostgreSQL, MySQL, or SQLite
- Ask questions in plain English
- Schema explorer & table-level access control
- Real-time results
Smart Chat Widget
A beautiful, embeddable widget with streaming responses. Customize colors, position, greeting, and suggested questions - no code required.
- Drop-in embed, iframe, or API
- Streaming responses with SSE
- Suggested questions & greeting messages
- White-label branding on Business
Team Inbox & Human Handoff
When the bot can't answer, it escalates to your team with full conversation context. Agents reply from the same dashboard - no separate chat tool needed.
- Unlimited team members on Pro and above
- Context-rich escalations
- Shared inbox for live chats
- Ticket system with categories & analytics
Analytics & Insights
Understand your support performance: deflection rate, CSAT, response times, and conversation source breakdowns.
- Deflection & resolution analytics
- CSAT tracking
- Usage meters & overage alerts
- Response source breakdown (AI vs human)
Enterprise Security
Security features built in from day one: end-to-end encryption, SOC 2 practices, SSO/SAML, RBAC, and audit logs.
- End-to-end encryption
- SSO/SAML on Enterprise
- Custom roles & permissions
- Audit log & session management
Lightning-Fast Streaming
Every response streams in real time with SSE. Multi-LLM routing and cost controls keep it both fast and cheap.
- Streaming responses with SSE
- Low-latency routing
- Multi-tenant isolation
- Scales from 10 to 200 requests/minute
Feature Questions, Answered
Which AI models does Chatonn support?+
Chatonn supports OpenAI GPT-4, Anthropic Claude, Google Gemini, Groq, and Ollama. You can bring your own API keys or use AI-managed responses where Chatonn handles the AI costs.
What is multi-LLM routing?+
Multi-LLM routing lets you assign different AI models to different conversation types - for example, a cheap model for routine questions and a powerful one for complex tickets. Auto cost-routing picks the cheapest capable model automatically.
Can Chatonn train on my own content?+
Yes. Chatonn's RAG knowledge base accepts PDFs, docs, FAQ pages, help-center exports, and website content. Every answer is grounded in your content with visible citations.
What is NL-to-SQL?+
NL-to-SQL lets customers ask natural-language questions about their data. The AI converts them to SQL queries against your PostgreSQL, MySQL, or SQLite database and returns results in real time.
Does Chatonn support human handoff?+
Yes. When the bot escalates, agents answer from the same dashboard with the full conversation history, so they never start from scratch.
See How It Works
Learn how to set up Chatonn, connect your AI providers, and train your knowledge base.