Elicana Is Available Today
Elicana compiles your document archive into a knowledge base you can query in plain language. It runs on your own machine, it reads the files you already have, and as of today anyone can sign up and download it.
I am quite excited to finally launch it for open beta.
What it does
You point it at a folder. It reads what is in there: PDFs, Word documents, spreadsheets, Markdown, an Obsidian vault, a Google Drive or OneDrive folder if you connect one. Then it compiles them.
Most tools index your documents and search them when you ask a question. Elicana processes the material first, making sense of it as a whole: connecting your corpus together, and reconciling the vocabulary that drifts across years of writing. So a search does not depend on you remembering which words you used in 2021, or on you tracking how your own terms shifted meaning since. After that you can ask questions across the entire corpus and get answers in plain English, grounded in your own material.
Why I built it
I did not set out to build a product. I had years of consulting material, meeting notes, MBA class notes, class material, key project decisions viewpoints, etc. spread across folders, and every time I wanted something out of it I was dragging the same files into a chat window and re-explaining the same context. The archive was the most valuable thing I owned and the least usable.
So I built something for myself, then rebuilt it, several times over. Somewhere in those iterations it started doing things I could not get from anything else I had tried. When I described the problem to other people, enough of them recognized it immediately that it stopped being a personal tool.
What you get
- Ingestion and compilation of your documents, in one binary, with no Docker and no setup project
- Hybrid search with plain-English synthesis, and answers grounded in your own material
- Google Drive and OneDrive as source connectors
- Local models through Ollama, or your own API key for a frontier model
- A guided first-run setup and a browser admin interface
- Up to two machines per subscriber
- Windows and Linux today, macOS coming soon
Where your data goes
Your source files stay exactly where they are, untouched. Elicana's output, the compiled knowledge base, stays on your machine. The app checks your license once a day, and that request carries key and entitlement data only. If you choose a frontier model, those queries go to your provider under your own API key, never through Elicana. You can switch between local and frontier models at any time.
What it costs, and what beta means
$20 a month, an early-bird rate for people taking a chance on something new. If that price changes later, I will email subscribers at least 30 days before it affects a renewal.
Beta here means rough edges and fast fixes, with support by direct email at support@elicana.ai. You will find bugs and I will fix them quickly. This is not a polished consumer app yet. It is a working tool for people whose documents genuinely need to be usable, who can live with some friction to get there.
Running it as a business rather than for yourself? Get in touch and we will talk about what you need.
Sign up
Register, add a card, and download at elicana.ai/get.
Related
- Why AI tools fail at your document archive
- I benchmarked plain RAG against corpus compilation on my own vault
- Nadella says you're paying twice. Do you know if you are?
Elicana is in open beta: no waitlist, no cohort. Get Elicana.