Edit the reply in English
Please send us the client version and confirm whether access works on the company network.
Translate your changes back into the customer’s languageZai-LLM* · AI add-on for Znuny
Zai-LLM adds AI features to your Znuny interface: summarise tickets, draft replies and find existing solutions. Add approved support tickets, internal knowledge articles or manuals to a provider such as Chat Data or Chatbase so your team can use company knowledge to handle customer enquiries.
Plug & Play · Your knowledge sources · Version updates included
* ZAI stands for “Znuny Artificial Intelligence”, LLM for “Large Language Model”. Zai-LLM connects these AI models to Znuny.

The Plug-&-Play AI add-on for Znuny
Simplified example using fictional data.
Hello,
Please check your VPN connection to the company network and try signing in again. If error E-104 persists, please send us a screenshot and the client version you are using.
Kind regards,
Your support team
Selecting text opens a new outgoing message with a draft reply for you to review and edit before sending. Illustration with fictional data.
Zai-LLM · ONE TICKET, STEP BY STEP
Follow one example to see how the modules work together. AI prepares information and draft replies. The agent reviews the suggestions, adds their expertise and decides what happens next.
Zai-LLM analyses the original customer request and prepares an internal initial proposal. With Chat Data attachment processing configured, supported images and documents can also be included.
The agent takes over the case and reviews the initial proposal.
CUSTOMER REQUEST
Since the client update, we cannot sign in from outside the company network. Access still works on site.
Attachment: login-error.pngCheck the VPN connection and client version. Use the difference between on-site and external access to narrow down the cause.
Alongside the AI proposal, ZaiQMatch suggests relevant, approved FAQ articles based on successfully resolved tickets. The agent checks which approach fits the current case.
The team can draw on existing experience without asking a colleague about every familiar issue.
RELEVANT FAQ SUGGESTIONS
A documented solution from a successfully resolved support case.
The guidance matches the symptoms. The next step is to check the VPN settings.
The agent can write a reply, request an AI draft directly or ask questions and give instructions in the interactive dialog. A selected error message can also provide the starting point. The draft can be transferred to an outgoing message for review and editing.
The agent decides what is sent to the customer.
OUTGOING MESSAGE · DRAFT
Hello,
Please check your VPN connection to the company network and try signing in again. If error E-104 persists, please send us a screenshot and the client version you are using.
Kind regards,
Your support team
When another agent takes over or a team lead needs an overview, Technical Summary brings together the ticket history, actions taken, results and outstanding points.
The summary supports further work and can be requested from within the ticket.
TECHNICAL SUMMARY
Once the issue is resolved, ZaiQGen prepares the solution as an FAQ draft. After review and approval, it becomes available for similar requests. If configured, draft creation can also be triggered automatically when the ticket is successfully closed.
The experience gained from this case can then help with the next similar request.
FAQ DRAFT FROM THE RESOLVED TICKET
Document the symptoms and the solution that worked in the ticket.
Review the content and approve the FAQ article. ZaiQMatch can then suggest it for similar requests.
The workflow shown requires the Zai-LLM Pro suite.
IN EVERYDAY BUSINESS USE SINCE 2024
Our AI integration for Znuny has been used in everyday business operations since 2024. We have continued developing it and adding useful, coordinated modules to Zai-LLM ever since. Experience from day-to-day ticket handling directly informs this ongoing work.
Our aim is to apply cutting-edge AI where it improves day-to-day ticket handling. We assess new features for their practical value and integrate them directly into Znuny workflows.
BUILT FOR EVERYDAY SUPPORT
Not every request is new. Yet research, handovers and drafting replies still take time every day. This is where Zai-LLM helps.
Handovers & escalations
Long histories take time to read. A technical summary brings the relevant findings together so the next agent can continue the case and consult the original messages when needed.
A useful starting point for the next person on the case.
Product illustration using fictitious example data.
Replies with context
AI Article Tasks helps draft replies and focused follow-up questions. Select a task, ask your own question or highlight a passage. Extended tasks use additional ticket context.
A draft your team can review and refine.
International support
AI Article Tasks provides an English translation of the request and a bilingual reply draft. Edit the English version, then use Backtranslate to update the customer-language reply. For document attachments, AI Document Translator also offers translation through DeepL or Google Cloud Translation.
Edit the English draft and translate your changes back into the customer’s language.
Bonjour, depuis la mise à jour, l’accès au portail échoue avec l’erreur E-104.
Since the update, portal access fails with error E-104.
Please send us the client version and confirm whether access works on the company network.
Translate your changes back into the customer’s languageMerci de nous indiquer la version du client et de confirmer si l’accès fonctionne sur le réseau de votre entreprise.
Review the result before sending the reply yourselfRecreated view with fictional data. English is the working language in this workflow. The button is illustrative only; this website does not run an AI request.
Product illustration using fictitious example data.
Knowledge for the whole team
After an agent successfully resolves a case, ZaiQGen can prepare an FAQ draft from the actions that worked. Once reviewed, the proven solution becomes available to the team. ZaiQMatch suggests relevant FAQ articles alongside AI proposals in similar cases, combining AI assistance with human-reviewed practical knowledge.
Useful solutions remain available for similar requests.
Cause documented. Solution checked.
The solution is available when a similar ticket arrives.
Product illustration using fictitious example data.
Zai-LLM · OUR RECOMMENDATION
Buy individual features for a specific task. If you want AI support across several stages, a suite costs less than buying all its features separately and lets your team use them together throughout the ticket workflow.
Prepare an FAQ draft from the solution that worked for the agent, then check it against the actual outcome.
Suggest reviewed FAQ articles with proven solutions in the sidebar of similar tickets.
Your team receives AI suggestions alongside relevant solutions from successfully resolved cases. The agent checks which approach applies to the current case.
Example of how a team can combine the features. Automation depends on configured functions and workflows; a suggestion does not become an unchecked customer reply.
Zai-LLM can draw on approved solutions from resolved tickets and FAQ articles when handling new enquiries. New team members can use experienced colleagues’ knowledge to prepare replies, without having to ask them about every familiar issue.
Zai-LLM · THE MODULES
Zai-LLM brings focused modules into your Znuny environment. See what each one does and how it helps your team.
From the first clue to the next check.
AI Initial Proposal creates an internal proposal from the original customer request. With attachment processing configured through Chat Data, supported documents and images can also be included. Agents receive suggestions for initial checks and focused follow-up questions.
Use cases & featuresAI Technical SummarySummarises the relevant history, previous actions and open questions. Useful for handovers, escalations and returning to older cases.
Use cases & featuresAI Solution ProposalDevelops a proposal from the case so far, helping your team plan further checks and guide the customer through the next step.
Use cases & featuresHelp customers move forward, across languages too.
Draft replies, ask questions and work with selected text directly on the article. Bidirectional translation helps you reply in the customer’s language: edit the English draft, then use Backtranslate to update the translated reply.
Use cases & featuresAI Document TranslatorPROTranslates supported document attachments through DeepL or Google Cloud Translation and returns the results to the ticket.
Use cases & featuresKeep useful solutions available to the whole team.
Prepares an FAQ draft from the steps that successfully resolved a support case. Your team reviews the article before making the proven solution available for similar requests.
Use cases & featuresZaiQMatchPROShows relevant FAQ suggestions in the ticket sidebar, bringing existing knowledge into the reply workflow.
Use cases & featuresAI ChatbotPROEmbeds a Chat Data or Chatbase chatbot in Ticket Zoom, giving agents access to the knowledge sources you maintain with that provider.
Use cases & featuresTHE PRODUCT TOUR
The full demo shows Zai-LLM in the real Znuny interface: article tasks, interactive questions, translation and more.
Watch on YouTubeWHAT YOU NEED TO START
Our recommended approach does not require you to operate an AI server. Setup follows the features your team chooses.
A maintained Znuny 6.5 LTS or Znuny 7 release with current security updates. We agree a suitable package release for your environment.
A business account with a supported online service. We recommend Chat Data for company knowledge, with Chatbase as an alternative. Direct AI tasks can use OpenAI.
Start with a few approved manuals or FAQs. Summaries without a company knowledge base do not require a document collection.
Try the chosen features with selected staff in your Znuny test environment first. We can help you configure the module and connect your AI provider on request.
AI SERVICES & COMPANY KNOWLEDGE
Our recommended approach connects Zai-LLM to a suitable online service, without an AI server of your own. Choose access for the features your team needs.
We recommend Chat Data for using your company knowledge with Zai-LLM. Ticket exports from Znuny, content from your company wiki, FAQs and documents describing technical issues can form a shared knowledge base. Your team can then draw on existing solutions when handling customer enquiries.
Chat Data supports configurable masking of personal data before it is sent to AI models and when conversation content is stored. Its Data Processing Agreement (DPA) documents data processing and provides for Standard Contractual Clauses (SCCs) where required. We can help you assess the settings and agreements for your intended use.
Another option for making company documents available for AI-assisted answers, including when your organisation already uses Chatbase.
Use your own business account for OpenAI’s interface (API) for summaries, reply drafts and other AI tasks. Usage is billed separately by OpenAI.
A ChatGPT subscription and API access are separate offerings.
Start with selected sources in Chat Data or Chatbase. You generally do not need to train your own AI model. Maintain the content and check answers against typical support questions.
DOCUMENT TRANSLATION
These services provide the dedicated translation features. Suitable API access is needed only if you use those features.
UNDERSTAND THE COSTS
Buy the module licence from Addonflow and pay the selected AI or translation provider directly. Optional setup and customisation are agreed separately.
AI assistance does not have to require a large budget. A suitable model and plan can keep running costs low. Costs depend mainly on usage, the amount of content processed and the provider you choose. We can help you estimate your requirements before you start.
We can integrate local AI models, for example through Ollama, OpenAI-compatible services and additional APIs on request. We review the interface and develop and test the required integration. Advice on model and hardware selection is available to suit your tasks, planned use and budget. The online services above remain our recommended starting point.
BEFORE YOU START
No. Zai-LLM extends your existing Znuny installation. Your team keeps working in the familiar interface and gains AI features within it.
We recommend a maintained Znuny 6.5 LTS or Znuny 7 release with current security updates. Before delivery, we match the package release to your environment. Assistance migrating from an older OTRS Community Edition to Znuny is available on request. Commercial OTRS installations are assessed separately.
No. You only need access to the services used by your chosen features. AI tasks can use Chat Data, Chatbase or the OpenAI API, for example. Document translation uses DeepL or Google Cloud Translation. Provider charges are additional to the module licence.
Agents can review and edit article reply drafts. Automatic workflows and any outgoing communication require separate configuration. We recommend starting with internal results and manual review.
Yes. We can assess individual requirements, queries and reports and quote for them separately. We agree on suitable changes based on your workflow.
You receive the module as an OPM package for installation through the Znuny Package Manager, together with its guide. Your offer specifies which updates and additional services are included.
Basic includes AI Initial Proposal, AI Technical Summary and AI Solution Proposal. Pro adds AI Article Tasks, AI Document Translator, ZaiQGen, ZaiQMatch and AI Chatbot. Enterprise provides the Pro features for an agreed deployment across multiple instances. See the package comparison for the full scope and prices.
Yes. Through Chat Data or Chatbase, you can make approved content from resolved support tickets, internal knowledge articles and manuals available as a knowledge base. The AI can draw on these sources for relevant requests. This is not the same as training your own language model. Agree internally which data may be used before setting it up.
ZaiQGen turns the resolution of a handled ticket into an FAQ draft. Your team checks that it accurately describes the solution that worked in practice and approves the article. ZaiQMatch suggests relevant FAQ articles for similar requests. Agents can then use reviewed, real-world solutions alongside AI suggestions. An AI draft is not automatically a validated article.
Yes. AI Article Tasks includes bidirectional translation: an English version of the customer request and a reply draft in English and the original customer language. After editing the English draft, use Backtranslate to update the customer-language reply. It is included in Pro and Enterprise or the standalone purchase of AI Article Tasks. Language support and quality depend on the AI model. Review the reply before sending.
A FIT FOR YOUR SUPPORT TEAM?
Tell us your Znuny version and the task you want to improve. We will show you the relevant modules and discuss what you need to get started.