SWX KI-Assistent Daniel Kreutz

SWX KI-Assistent is a local RAG (Retrieval-Augmented Generation) system for Windows designed to help users search and retrieve information from their own documents without relying on cloud services or external servers. It supports importing PDFs (including scanned documents via built-in OCR), Word, Excel, text files, images, and OTRS ticket exports, enabling users to find content by meaning or keywords and receive concise AI-generated answers with source citations—all while maintaining complete data privacy.

Key Features:

• Imports various document types, including PDFs, Word, Excel, text files, images, and OTRS ticket exports.
• Searches documents by meaning in addition to keyword searches, ensuring accurate retrieval of relevant information.
• Generates concise answers from imported documents using a local language model with source citations.
• Runs entirely offline after setup, ensuring data privacy and no reliance on external servers.
• Supports multiple local language models for different performance needs, allowing users to switch as needed.
• Organizes document collections into separate "knowledge spaces" to keep topics distinct during searches.

Target Audience and Benefits:

Ideal for organizations that prioritize secure document management and AI-driven search without cloud dependencies. SWX KI-Assistent is perfect for public authorities, companies with sensitive data, IT departments, and small teams seeking an offline AI solution. It streamlines knowledge retrieval from internal documents, reducing the time spent searching for information and improving efficiency.

Available via Winget, SWX KI-Assistent can be easily installed or deployed using the Winget command line, ensuring smooth integration into environments that value privacy, security, and local processing of data.