OpenWebUI

With OpenWebUI, you can flexibly change models, save and resume chats, upload files and create your own knowledge databases that can be accessed directly in the chat context. Web search, predefined prompts and customised skills are also available.

AI chats & model management

• Intuitive chat interface: Looks similar to ChatGPT, but offers much more control. You can organise, share and export chats.
• Multi-model support: You can use different language models (e.g. gemma-4, qwen3.6) in parallel and even compare them directly in a chat.
• Voice control: With the integrated "Text-to-Speech" and "Speech-to-Text" function, you can convert voice messages into written text and vice versa. In combination, you can even talk to the chatbot.
• Functionality: You can create predefined (long) prompts or define skills. This makes it easy to store and reuse recurring tasks that require a detailed prompt.

Integrated RAG engine (knowledge database)

The absolute highlight: OpenWebUI has an in-built RAG system.

• Document upload: You can upload PDFs, text files, Word documents or URLs directly into OpenWebUI.
• Vectorisation: the tool automatically splits documents, converts them into Mathematics vectors and stores them in an integrated vector database on university servers.
• Knowledge-based answers: When a user asks a question, the system first searches your uploaded documents for the appropriate answer and passes this information on to the local AI model. The AI therefore only answers based on your content.