Local-first AI · live today

Own your intelligence.

The local-first AI platform — your knowledge, your learning, your machine. Nothing leaves your laptop unless you allow it.

Free, open, and yours. Knowledge engine + learning app, live today.

$ pip install docnest-ai

>>> from docnest.reader import UDFIndex
>>> idx = UDFIndex.load("report.udf")
>>> r = idx.query("What was Q3 revenue?")

>>> r.answer        # "Q3 revenue was $38M, up 22% YoY."
>>> r.tokens_used   0   # answered from structure, no LLM

Answered straight from the document — no model call, no token bill.

Built on DOCNEST engine Knovex app Open .udf format On-device. Always.

Most AI reads your private documents on someone else's computer, charges you to re-read them every time, and forgets it all when you close the tab.

The platform

Four layers, one foundation

Knowledge, understanding, agents, and models — stacked like hand-laid stones. We lead with what's live and mark the future as the future.

01 Live
Knowledge

DOCNEST

Turns any document into portable, structured .udf knowledge you own.

02 Live
Understanding

Knovex

A desktop app to chat, summarize, search, and actually learn from your knowledge.

03 Roadmap
Agents

Keepvidya Flows

Agentic workflows that act on your knowledge, where it lives.

04 Roadmap
Models

Keepvidya Core

Small LLMs running on-device — no cloud required.

Local-first foundation — your machine, your data. Always.
Products

Tools you run yourself

Each one open, each one yours — with one real proof.

Why local-first

The private alternative to cloud AI

Same task, two very different deals.

What matters
Cloud AI
Keepvidya
Your data
Cloud Uploaded to someone else's computer.
Local-first Stays on your machine.
Cost to re-read
Cloud Re-billed every time it re-reads your docs.
Local-first Embed once; ask freely.
What you keep
Cloud Chat-and-forget when the tab closes.
Local-first Learn it and keep a .udf you own.
The format
Cloud Proprietary lock-in.
Local-first Open, portable format.
From the blog

Notes from building it

Field notes on local-first AI, written while shipping it.

GitHub · Source

Follow the work in the open

Every layer of the platform is built in public. Star the repos and follow along.

Browse the repos →

Own your intelligence.

Start with the app that's live today. Your knowledge, your learning, your machine.