Find clear and concise answers to frequently asked questions about our AI solutions and integrations.
INNOCHAT is an AI chatbot builder that allows you to create intelligent conversational assistants for your business. It connects to your data and uses it as a knowledge base to answer queries. You can upload data directly, import it from cloud storage partners such as Google Drive, Dropbox or OneDrive, specify URLs for automatic content scraping, or enter text directly. The chatbot learns from your specific content to provide accurate and relevant answers. You can embed the chatbot on your website or integrate it into various platforms such as Slack, WhatsApp, Zapier, Make, and more. InnoChat is an AI chatbot builder that allows you to create intelligent conversational assistants for your business. It connects to your data and uses it as a knowledge base to answer queries. You can upload data directly, import it from cloud storage partners such as Google Drive, Dropbox or OneDrive, specify URLs for automatic content scraping, or enter text directly. The chatbot learns from your specific content to provide accurate and relevant answers. You can embed the chatbot on your website or integrate it with various platforms such as Slack, WhatsApp, Zapier, Make and more.
Our SaaS solution is operated securely in the cloud, hosted by Amazon Web Services (AWS) in the Frankfurt region (EU-CENTRAL-1), in compliance with the GDPR. Your data is protected by encryption during transmission and storage, and regular automatic backups are performed. Access is strictly controlled and only possible for authorised INNOQ employees. There is no third-party access. We comply with the requirements of the Swiss Data Protection Act (revDSG) and the EU GDPR.
INNOCHAT supports over 95 languages thanks to the integration of leading AI models such as GPT-4, Claude 3, Gemini Pro and DeepSeek. GPT-4 offers the broadest language coverage, Gemini is strong in rare languages and OCR, and Claude impresses with its quote accuracy. DeepSeek is optimised for Chinese-language content. For multilingual texts and documents, we recommend GPT-4 or Gemini.
INNOCHAT supports text, Office, PDF, CSV, JSON, XML and code files as well as publicly accessible websites. Google Drive, Dropbox, OneDrive and YouTube links (with transcription) are also possible. PDFs up to 50 MB, Office files up to 25 MB, total limit 10 GB per chatbot. Web content is captured once; automatic updates are not active.
INNOCHAT uses LLMs from OpenAI (GPT-3.5, GPT-4o, GPT-4.1, GPT-o1/o3), Anthropic (Claude 3/3.5/4), Google (Gemini 1.5, 2.5) and DeepSeek (R1, V3). They cover different contexts from 2K to 1M tokens. GPT-4o-mini, Claude Haiku and Gemini Flash are suitable for quick answers. For complex content, we recommend Claude Opus, GPT-4.1 or Gemini 2.5 Pro.
Yes, you can share chatbots via link, iFrame, widget or API. innoChat allows team access with roles (observer, editor, admin), public links with access control, website embedding or sharing with other account holders. Usage and sharing can be managed via the dashboard and revoked at any time.
Yes, you can customise your chatbot's behaviour – such as tone of voice, area of expertise, response style or personality. Templates such as support agent or sales assistant are available, as are your own rules. Adjustments take effect immediately, without the need for retraining.
You can easily embed INNOCHAT into your website using JavaScript or iFrame. Ready-made connectors are available for popular platforms such as Shopify, WordPress, and Confluence. Individual systems can be integrated via the REST API. For automated processes, we recommend Zapier, Make, or n8n – no programming required.
AI agents are specialised modules within your chatbot – e.g. for support, sales or analysis. Each agent uses a suitable AI model and follows its own rules. You create and control agents without programming. An AI supervisor automatically recognises which request suits which agent and ensures a seamless conversation flow.
Message Credits (MC) measure token usage in your chatbot conversations. Different AI models consume different amounts of credits per response. For example, GPT-3.5 is the most economical, while GPT-4 is the most powerful. You can view consumption in the dashboard and optimise credits specifically for models, response length and context.
Feature Credits activate advanced functions in innoChat – e.g. email notifications, web searches, PDF creation or API calls. Each action consumes a defined number of FCs. You only pay for the additional functions you use and can keep track of your consumption in the dashboard.
Tokens are the smallest units of language that AI models process. The context size determines how many tokens are considered at the same time – e.g. questions, conversation history and documents. Smaller contexts (4K) are sufficient for simple queries, larger ones (128K–1M) for complex analyses. The more context, the more accurate the answer – but it is also more resource-intensive.
The storage space includes documents, chat histories, analyses and media. 1 GB is sufficient for around 5,000 pages or 100,000 conversations. Small businesses often manage with 10 GB, while larger teams need more. Your storage usage is visible in the dashboard. Additional space can be added at any time.
Collaborators are team members with individually assigned rights for your chatbots. Roles such as viewer, editor or administrator can be customised for each chatbot – e.g. for support, content or analysis. This allows teams to work together securely and purposefully, without unnecessary privileges.
Function calling allows your chatbot to perform tasks such as checking order status, booking appointments or executing API calls – directly in the conversation. You connect external systems, define functions in the dashboard, and the chatbot automatically calls them up as needed. This turns the bot into an active helper instead of just a responder.
Variables store information during a chat session, such as name, language, product selection or order number. This allows the chatbot to respond in a personalised manner and maintain context. There are user, session and system variables that can be used dynamically in the conversation.
Automatic tags classify chat histories in real time according to topic, urgency or outcome. The AI recognises patterns such as complaints, queries or successes and assigns appropriate labels. This allows conversations to be filtered, evaluated and forwarded in a targeted manner – without manual tagging.
AI Supervisor Override enables fixed rules for forwarding enquiries – regardless of AI logic. For certain keywords, customer characteristics or times, a defined agent is automatically activated, e.g. for ‘cancel subscription’, retention support is activated directly. This keeps your routing reliable and controllable.
Your security and privacy are important to us. We implement strict measures to prevent data loss, unauthorised access and unauthorised use. Access to your data is restricted to authorised employees who are bound to confidentiality.
Our systems are ISO, SOC and GDPR certified and offer the highest level of protection. We use encryption technology and adhere to the Critical Security Controls of the Centre for Internet Security. Strict regulations and procedures are in place, supported by comprehensive security training for our employees.
INNOQ only uses your data to provide and improve your chatbot. Your files, conversations and settings are processed securely – never used for advertising or passed on to third parties. Everything remains under your control and can be deleted at any time.
No, neither INNOQ nor the affiliated providers of language models such as OpenAI, Anthropic, Google or DeepSeek use your data to train their models. All information transmitted is used exclusively for processing individual enquiries and is not stored. Your data remains confidential and under your control.
No. innoChat is a cloud-based platform and does not require any local installation. It is used entirely via a web browser. Access is platform-independent and works on desktop and mobile devices.
Use in the browser
No, unless a support case makes access absolutely necessary and access is explicitly permitted.
INNOQ does not access your uploaded documents as a matter of principle. Access by authorised technicians only occurs if a specific technical problem, a support request or a security-related situation requires it. Even in such cases, strict access guidelines and comprehensive control and protection measures apply.
All access is limited to what is necessary, is logged and ends automatically after the analysis or troubleshooting is complete. You retain control over your data at all times.
No, full end-to-end encryption is not possible in innoChat, as the AI needs to analyse your content in order to generate responses. However, your data is encrypted during storage (AES-256) and transmission (TLS 1.3), stored in strictly separate locations and protected by GDPR-compliant security measures.
innoChat protects your data with AES-256 encryption, TLS 1.3, access controls, daily backups and 24/7 threat monitoring. Our infrastructure is GDPR and CH-DSG compliant and is based on secure AWS servers in the EU. External audits, security policies and an incident plan ensure the highest level of cloud security.
No, only small text excerpts (chunks) are sent to LLM providers such as OpenAI or Google for response – never entire documents or personal data. Your content remains confidential, is not stored and is not used for model training. Processing is carried out securely via our AWS servers in Frankfurt.