Features
Everything you need to chat with your knowledge
Upload anything
PDFs, notes, and web articles are supported.
Chat with your docs
Ask questions and get concise, cited answers.
Semantic search
Vector similarity finds the most relevant chunks.
Private by default
Your data stays in your workspace with access controls.
How it works
Retrieval-Augmented Generation in four steps
Upload
Add your PDFs, notes, and articles.
Chunk & embed
We split docs and create vector embeddings.
Store & retrieve
Indexed in Postgres with pgvector for fast similarity search.
Answer
Relevant context is fed to the LLM to craft a concise answer.