Qnaps has announced a major update to its core file search application, Qsirch, introducing the new 'Qsirch AI Mode'. This advanced feature uses a Vision Language Model (VLM) and a Large Language Model (LLM) to enable users to locate desired files among documents, images, video and audio recordings by describing them in simple, everyday language.
The AI mode provides advanced semantic understanding, enabling seamless search across several main categories. Document search supports multiple formats, including PDF, DOCX and PPTX. The AI understands contextual meaning, relying not just on exact keyword matching, so users can find specific paragraphs based on broader context.
The image search experience is further simplified by local VLM inference. Users no longer need to create albums in advance or add tags manually; simply describing the visual scene is enough, as the AI analyses the photo content and returns accurate results immediately.
For video (MP4, MOV) and audio (WAV, MP3, M4A) files, the system introduces bidirectional synchronization between the timeline and the transcript. Users can locate specific moments in recordings based on remembered dialogue, jumping directly to that point by clicking on the text, without the need to manually search through long recordings.
Qsirch AI Mode utilises the computing power of the QN NAS. By performing all LLM and VLM inference locally on the device, the solution preserves the confidentiality of sensitive documents and eliminates the costs associated with transferring data to the cloud. Completely independent of external networks, the core AI mechanism maintains high search performance within corporate internal networks and strictly isolated environments.
Security is further enhanced by integration with existing NAS permissions for shared folders, strictly enforcing user access levels and minimising the risk of data exposure. This advanced functionality provides intelligent data management and enterprise-level protection for professional users. The new Qn 7.1.0 is now available to download from the Qn App Centre.
