Yes. Everything runs locally in your browser. Files are parsed in a Web Worker, embeddings are computed on-device, and the index is stored in your browser's Origin Private File System (OPFS) - accessible only by this site.

What file types are supported?

Plain text (.txt, .md, code files), PDF, DOCX, and images (PNG, JPG, etc. via OCR). Images require an initial OCR pass before embedding.

Will my index survive a page refresh?

Yes. The vector index is stored in OPFS and persists across sessions. You can clear it at any time from the Index Manager.

How big can my document collection be?

Up to 500 files per ingestion batch. The index size is limited by your browser's OPFS quota (typically ~10% of available disk space).

Personal RAG - Private Semantic Search

Upload documents and search them semantically using local AI embeddings. Runs 100% in your browser - files never leave your device.

How to use

1
Upload your documents
Drop files or folders - supports .txt, .md, .pdf, .docx, code files, and images (via OCR).
2
Wait for indexing
Documents are chunked and embedded locally using all-MiniLM-L12-v2. The 33MB model downloads once and is cached.
3
Search semantically
Type a question or description. Results are ranked by semantic similarity - no keyword matching needed.

Why use this tool?

Files never leave your device - all processing happens in the browser via WebAssembly
Index persists across sessions using OPFS (browser's private file system)
Supports PDF, DOCX, images (OCR), and plain text files
True semantic search using neural embeddings - finds meaning, not just keywords

Frequently asked questions

Is my data private?: Yes. Everything runs locally in your browser. Files are parsed in a Web Worker, embeddings are computed on-device, and the index is stored in your browser's Origin Private File System (OPFS) - accessible only by this site.
What file types are supported?: Plain text (.txt, .md, code files), PDF, DOCX, and images (PNG, JPG, etc. via OCR). Images require an initial OCR pass before embedding.
Will my index survive a page refresh?: Yes. The vector index is stored in OPFS and persists across sessions. You can clear it at any time from the Index Manager.
How big can my document collection be?: Up to 500 files per ingestion batch. The index size is limited by your browser's OPFS quota (typically ~10% of available disk space).

Related tools

Local Vector Embeddings Tester (ONNX)

Compute cosine similarity locally using an ONNX embedding pipeline. Zero delays.

RAG Document Chunking Preview

Simulate and test text chunking algorithms and overlap characters instantly.

Popular right now

JSON Formatter & Validator

Format, validate, and minify JSON instantly in your browser. Your data never leaves your device.

JWT Decoder

Decode JWT tokens and inspect header and payload instantly in your browser. Your tokens never leave your device.

Word Counter

Count words, characters, sentences, and estimate reading time instantly in your browser. No sign-up required.