# Open Ticket AI > Private, on‑premise AI for ticket automation. Use Full On-Prem as the primary offering. Free Cloud Trial is for evaluation only and is not the production deployment model. **Open Ticket AI** OTAI is private AI for your helpdesk — routing, priorities, summaries, and more. Tuned to your structure. Full On-Prem first. Everything runs on your infrastructure. Capabilities: - Multi-Attribute Classification: Classify routing target, priority, and other single-select fields from your helpdesk setup — so tickets land with the right team and priority without exporting ticket payloads. - Ticket Summary: Short, consistent ticket summaries for agents — written back into your ticket system so people see context without reading every reply. - Rock-Solid Install: Deploy with Docker next to your ticket system. Connectors for Zammad, OTOBO, Znuny and more read new tickets locally and write results back — no ticket export required. - Title Rewrite: Suggest clearer ticket titles from the conversation. Included in the product; we keep it out of the commercial headline until quality and docs match. - Field Extraction: Pull structured values into ticket fields from the conversation. Included in the product; not a lead claim until quality and docs match. - Reply Drafts: Suggested agent replies grounded in your knowledge — coming soon, once they survive a real ticket-system demo without babysitting. - Chatbot: On-prem chatbot for internal or FAQ-style answers — coming soon, once quality is solid in a live ticket-system demo. Deployment variants: - **Free Cloud Trial** (Free) — Evaluate & test - Routing, priority, multi-field AI + summaries - No field-count limit - Hosted Free Cloud Trial — runs online, not on your servers - Built to evaluate before Full On-Prem - Community support - Availability: Available Now - **Full On-Prem** (From ~€500 / month) — On request · agent-based software - Routing, priority, multi-field AI + summaries - Unlimited values per field - Quote depends on agents + pack (Entry / Standard / Premium) - Run on your own GPU — bring your own or optional appliance - Requires a GPU with at least 24 GB VRAM - Fully isolated — nothing leaves, not even setup data - Maximum data sovereignty & compliance - Support & updates included - Availability: Requires customer GPU Deployment highlights: - ON YOUR SERVERS - GPU REQUIRED - NOTHING LEAVES - BUILT FOR HELPDESKS ## Products - [Open Ticket AI](https://openticketai.com/en/product/): OTAI is private AI for your helpdesk — routing, priorities, summaries, and more. Tuned to your structure. Full On-Prem first. Everything runs on your infrastructure. ## Solutions - [Open Ticket AI for Zammad](https://openticketai.com/en/solutions/zammad/): Train a custom AI model for your Zammad groups and other ticket attributes. Full On-Prem training, on-prem inference — no ticket data leaves your infrastructure. - [Open Ticket AI for OTOBO](https://openticketai.com/en/solutions/otobo/): Train a custom AI model for your OTOBO queues and other ticket attributes. Full On-Prem training, on-prem inference. - [Open Ticket AI for Znuny](https://openticketai.com/en/solutions/znuny/): Train a custom AI model for your Znuny queues and other ticket attributes. Full On-Prem training, on-prem inference. - [Open Ticket AI for Government](https://openticketai.com/en/solutions/government/): Full On-Prem AI classification for citizen requests — train and run on your infrastructure, GDPR-aligned. ## Services - [Internal Chatbot](https://openticketai.com/en/contact/): Available as a Service · On Request - [FAQ Suggestions](https://openticketai.com/en/contact/): Available as a Service · On Request - [Agent Answer Drafts](https://openticketai.com/en/contact/): Available as a Service · On Request - [FAQ Chatbot](https://openticketai.com/en/contact/): Available as a Service · On Request - [Basis](https://openticketai.com/en/contact/): Support & maintenance subscription add-on for standard coverage - [Premium](https://openticketai.com/en/contact/): Support & maintenance subscription add-on with premium SLA coverage - [Standard Stack](https://openticketai.com/en/contact/): Evaluation & Testing - [High Volume Training](https://openticketai.com/en/contact/): Production Ready - [Complex Use Cases](https://openticketai.com/en/contact/): Enterprise Strategy - [Fine-tuned Model](https://openticketai.com/en/contact/): Custom AI model training extension - [Evaluation Report](https://openticketai.com/en/contact/): Comprehensive performance metrics - [System Tuning](https://openticketai.com/en/contact/): Platform-specific format optimization - [Ongoing Updates](https://openticketai.com/en/contact/): Periodic dataset refreshes - [OTAI Automation Pack](https://openticketai.com/en/contact/): Implementation of 3 core automations - [Custom Model](https://openticketai.com/en/contact/): Synthetic Training Focus - [Custom Model](https://openticketai.com/en/contact/): Real Data Training - [Custom Development](https://openticketai.com/en/contact/): Special Requirements ## Documentation - [Docs home](https://openticketai.com/en/docs/): Technical documentation for OTAI. - [OTAI Studio documentation](https://openticketai.com/en/docs/otai-studio/): Handbook for installing, configuring, and operating on-prem OTAI Studio. - [Zammad Commander: Internal AI Chat for Your Zammad Helpdesk](https://openticketai.com/en/docs/zammad-commander/): Zammad Commander is an open-source internal admin chat for Zammad. Run LibreChat plus the Zammad MCP server to search tickets, summarize threads, and draft replies — in the cloud or fully on-prem with Ollama. - [OTOBO Python SDK](https://openticketai.com/en/docs/otobo-python-sdk/): Async Python client for the OTOBO GenericInterface REST API — install with pip install otobo. - [Zammad MCP Server: Open Source Setup for Claude and Cursor](https://openticketai.com/en/docs/zammad-mcp-server/): Install and use the open-source Zammad MCP Server for Claude, Cursor, and MCP clients with typed ticket tools, safe access control, and Zammad API token setup. - [OTAI Runtime](https://openticketai.com/en/docs/otai-runtime/): On-prem ticket system connectors for OTOBO, Znuny/OTRS, and Zammad - [Znuny Python SDK](https://openticketai.com/en/docs/znuny-python-sdk/): Async Python client for the Znuny GenericInterface REST API — install with pip install znuny. ## Optional - [Home](https://openticketai.com/en/): Private, on‑premise AI for ticket automation. - [Contact](https://openticketai.com/en/contact/): Talk to sales about Full On-Prem. - [Blog](https://openticketai.com/en/docs/blog/): Product updates and helpdesk AI articles. - [On-Prem Ticket AI: GPU Hardware You Need and What It Really Costs](https://openticketai.com/en/docs/blog/on-prem-ticket-ai-gpu-hardware-economics/): What GPU VRAM you need for on-prem ticket AI, how hardware and API costs compare, and what Open Ticket AI runs locally — classification, summaries, Full On-Prem. - [Znuny-LLM: A Practical Guide to Self-Hosted AI for Znuny [2026]](https://openticketai.com/en/docs/blog/znuny-llm-self-hosted-ai-guide/): Explore Znuny-LLM features, deployment choices, security controls, and rollout considerations for service desks evaluating self-hosted AI. - [Best Ollama Model for Zammad AI: On-Premise LLM Guide [2026]](https://openticketai.com/en/docs/blog/best-ollama-model-zammad-ai/): Which Ollama model works best for Zammad 7 AI? A practical, on-premise model guide — the structured-output gotcha, real speed numbers, VRAM sizing, and which local LLM to pick for summaries, title rewriting, and classification. - [Zammad Commander: An Open-Source Internal AI Chat for Zammad](https://openticketai.com/en/docs/blog/zammad-commander-internal-ai-chat-for-zammad/): Build an internal admin chat for your Zammad helpdesk with LibreChat and the Zammad MCP Server. Search tickets, summarize threads, and draft replies — in the cloud or fully on-prem with Ollama and Qwen 3.6. - [Zammad MCP Server: Setup Guide for Claude and Cursor](https://openticketai.com/en/docs/blog/zammad-mcp-server-setup-and-usage/): Set up the free Zammad MCP Server for Claude, Cursor, and other MCP clients. Install with uvx, connect your Zammad API token, and use 30+ helpdesk tools. - [ROI Calculator](https://openticketai.com/en/roi-calculator/) - [Documentation](https://openticketai.com/en/docs/)