What is the AI Knowledge Base?

It lets you upload multiple documents and ask questions across all of them — like a private, local version of NotebookLM with no cloud or data sharing.

How many documents can I add?

You can index multiple PDFs and text files. All indexing and Q&A happen locally in your browser.

What is the AI Knowledge Base?

It lets you upload multiple documents and ask questions across all of them — like a private, local version of NotebookLM with no cloud or data sharing.

How many documents can I add?

You can index multiple PDFs and text files. All indexing and Q&A happen locally in your browser.

🗂️

Personal Knowledge Base

Build a searchable knowledge library from multiple documents. Auto-tags, topic clustering, and semantic search across all your files — 100% local.

100% LOCALMulti-PDF supportSemantic searchLocal storageDocument management

📂

1. Upload your documents

Upload PDF, TXT, CSV, or Markdown files to build your knowledge base. Documents are parsed, chunked, and indexed using TF-IDF vectorization — all locally in your browser.

🔍

2. Choose search or Q&A mode

In Search mode, find relevant passages by keyword. In Q&A mode, the local AI reads the top results and generates a synthesized answer with source citations.

📝

3. Ask a question or search

Type your query and press Enter. The system ranks all text chunks by relevance using cosine similarity and returns the best matches.

📋

4. Review & export results

See matching passages with relevance scores. In Q&A mode, get an AI-generated answer with source citations. Copy results for your notes.

💡 Tips

• Q&A mode requires an AI model download (~200MB) on first use. Search mode works instantly without any download.

• The knowledge base uses TF-IDF (Term Frequency-Inverse Document Frequency) for semantic matching — no data leaves your device.

• Upload multiple documents to build a comprehensive knowledge base. You can remove individual documents at any time.

• For best results, upload well-structured documents with clear paragraphs and headings.

🧠