This will scan the NOMAD's storage directories for any new files and queue them for processing. This is useful if you've manually added files to the storage or want to ensure everything is up to date. This may cause a temporary increase in resource usage if new files are found and being processed. Are you sure you want to proceed?
{aiAssistantName} Knowledge Base Integration
When you upload documents to your Knowledge Base, NOMAD processes and embeds the content, making it directly accessible to {aiAssistantName}. This allows{' '} {aiAssistantName} to reference your specific documents during conversations, providing more accurate and personalized responses based on your uploaded data.
Enhanced Document Processing with OCR
NOMAD includes built-in Optical Character Recognition (OCR) capabilities, allowing it to extract text from image-based documents such as scanned PDFs or photos. This means that even if your documents are not in a standard text format, NOMAD can still process and embed their content for AI access.
Information Library Integration
NOMAD will automatically discover and extract any content you save to your Information Library (if installed), making it instantly available to {aiAssistantName} without any extra steps.
Auto-index new content for AI?
Indexed content typically uses 5–10× the original file size on disk. Changes apply to new content added after this setting changes.
This will re-process every document currently in your knowledge base — about {storedFiles.length} file{storedFiles.length === 1 ? '' : 's'}. For each file, NOMAD will delete the existing embeddings from Qdrant and queue a fresh embedding job using the current chunking and embedding model.
What this is for
Use this when the embedding model or chunking logic has changed, or when you suspect stored vectors are stale. Files on disk are not deleted, and any orphan points whose source file is no longer present will be preserved untouched (see Reset & Rebuild if you want a fully clean slate).
Heads up
This will permanently delete every point in the
nomad_knowledge_base Qdrant collection and rebuild from the
{storedFiles.length} file{storedFiles.length === 1 ? '' : 's'} currently
on disk. The collection is dropped, recreated, and every file is re-queued for embedding.
How this differs from Re-embed All
This action is destructive and cannot be undone
Type RESET exactly (uppercase, no spaces) to enable the confirm button.
)}This will delete the existing embeddings for{' '} {confirmReembed.displayName} and queue a fresh embedding job. The file on disk is not touched.
Heads up