The flagship · local AI you run · v3.0.3
Stop renting AI.
Run it on your own machine.
A model you choose. A core that never changes under you. A signed report of every outside connection OYA made in the session: evidence you can check, not a certificate — and every tool in this store connected to it.
Software that answers to you: you run it on a model you choose, it counts every outside connection, and you connect it to every Ownware tool. Buy once. The license never expires.
Runs on: Node.js 22+; one command (`oya up`) starts the engine and the web app; a local model server of your choice; Linux/macOS (Windows via WSL).
See it running
The product
Every shot below is the app itself, taken from the public showcase — the same one you can open. The runs it replays were recorded on a desktop with a local Qwen3.8-27B model, and the strip along the bottom is live in each one. Click any shot to open it full size.
Pricing — one time, yours forever
What this license covers
- Install it on the machines you personally use
- Full source — read it, change it, keep it
- Use it for your own paid work
- Not a company-wide rollout — that is the Business license
What this license covers
- Roll it out to everyone in your organization
- Everything the Personal license grants
- Client engagements outside your organization need the Agency license
What this license covers
- Everything the Business license grants
- Use it inside client engagements, on their material
- The source stays yours to run, never yours to resell as a product
No subscription · no meter · never expires · source included · frozen core
Version 3.0.3 · instant download, license key included · refunds handled case by case through Lemon Squeezy — we’d rather refund than argue · the offline test suite ships in the box (854 automated checks pass on this version · see the run) · if we ever stop publishing, after 3 years your copy relicenses to MIT — it’s in the license. The complete wording is on the terms page. Or the whole store with it: Ownware OS + Own Your AI — $1,499.00.
Evidence, not promises
Watch it work. Then check its receipt.
oya report --verify — the report, and how to verify it ↗The transcript is a replay against Invora’s sample data — the showcase replays recorded runs, and there is no model on that box. The report beside it is the real sample published on the showcase, generated 18 August 2026 and signed with Ed25519; your own copy produces the same report about your own session with oya report --airplane --pdf.
What you get
One engine. Everything on your side of the wire.
Any OpenAI-compatible local server — llama.cpp, Ollama, LM Studio, vLLM. Recommends a model that fits your hardware and downloads it, verified. Or point it at an endpoint you choose; the app then says REMOTE, in red, because that is what it is.
The OUT channel shows how many connections left your machine this session and to which hosts — live. Airplane mode blocks OYA's outside connections; sealed mode runs shell and scripts with no network so the count is the whole number. Prove it → a signed report you can hand to anyone.
On your instruction it plans, calls tools and acts — files, shell, code, search, pages, PDFs, spreadsheets, calendar, mail, SQL — with a live one-line status while it works and the full trace afterward. Mutating actions ask first, unless you say otherwise.
Read, write, diff, search files; run tests; import CSV/XLSX into SQL; read and write .docx/.xlsx/PDF; redact personal data; calendar .ics; IMAP/SMTP mail over TLS; a private web search through your own SearXNG. Then add pieces — MCP servers, plugins, skills — each showing where its data goes.
Every product in this store ships /mcp. Paste an API key from the product's API-settings page and it reads your invoices, rosters and work orders and answers in plain words — walked end to end against live products, read-only, with the transcripts.
A first-class browser app (the Mixer: sessions, pieces, files, knowledge, skills, ledger, prove-it) and a full CLI with sessions, scheduling, blueprints, cron and a signed attestation chain. Both talk to the same local engine.
The core is finished software: SCOPE-FREEZE.md fixes what it is; an offline test suite keeps it there; security fixes and honest maintenance only. You will not wake up to a different product.
Vanilla Node.js and ES modules, no build step, no third-party runtime dependencies (the supply chain is derived and stated in the report). Read it, change it, run it for as long as you like.
Runs on: Node.js 22+; one command (`oya up`) starts the engine and the web app; a local model server of your choice; Linux/macOS (Windows via WSL).
What the same job costs rented
Every figure below is from the vendor’s own pricing page, with the date it was read.
$20/month billed monthly; $17/month with the annual discount ($200 up front). A free tier exists; Team seats are $20 per seat/month billed annually ($25 monthly)
From $100/month — the 5x usage tier; the 20x tier costs more
Business Standard with Copilot $23.50 user/month paid yearly; Business Premium with Copilot $32.00 user/month paid yearly
Those months assume a machine you already own: as a guide, a 24 GB graphics card can run a quantized 27B model, and the setup recommends a model sized for your hardware.
The full table — every competitor, what each does better than us, and who should stay with them — is on the comparison page. Nothing here is ordered by commission.
Before you buy
Is Own Your AI right for you? Not for everyone — deliberately.
- Not a model: you bring the model server (llama.cpp, Ollama, LM Studio, vLLM); the setup recommends and downloads one, verified
- Linux and macOS are the execution hosts; Windows users run it under WSL (the isolation verdict is self-verifying there)
- No desktop app, editor plugin, image generation, speech, browser automation or messenger bridges in the core — those are pieces, or not at all
- The public showcase at oya.ownware.io replays recorded runs; there is no model on that box — OYA runs on yours
- Mail is IMAP/SMTP over TLS with app passwords; OAuth-only mailboxes are not supported
- If it fits none of that, the tools in this store work without it — and keep working if you never buy it.
Still the right fit? See the price → · Not sure — see who should stay with the cloud assistants →
After you buy
Your download link always serves the current build at the time you fetch it — re-download any time from your order page or the buyer portal. No renewal fee to keep what you bought.
Email support for installation and for defects in the code as delivered — a person answers, usually within a day or two. It does not cover custom development or server administration. What support covers →
Refunds are handled by Lemon Squeezy as Merchant of Record, case by case — if something is broken and we can’t fix it, we’d rather refund than argue. EU consumers additionally hold the statutory 14-day right until delivery starts. Refund terms →
Nothing stops. You hold the full source — Node.js you can read and change — it runs on your own machine, and the software does not phone home: there is no license check that can fail and no cloud that can be switched off. LICENSE.txt ships inside the zip, so your right to keep running it never depended on this site existing.
Questions about Own Your AI
What do I need to run it?
A machine with Node.js 22+ and a local model server (llama.cpp, Ollama, LM Studio or vLLM). A 24 GB GPU runs a strong 27B model; 8–16 GB runs the smaller ones the setup recommends. Linux and macOS are the execution hosts; Windows runs it under WSL.
Does it phone home?
No. The only connection made on your behalf without asking is to the model endpoint you configure. Everything else is a tool you invoke, counted in the ledger. The report lists every host that appears in the source and its stated purpose; the tests fail if a host is unclassified.
Is a model included?
No — models are licensed by their publishers and you pick the one that fits your hardware. The setup recommends one and downloads it, verified. You can also point OYA at any OpenAI-compatible endpoint you trust; the app then labels the model REMOTE.
How is this different from a chat UI like Open WebUI or a cloud assistant?
It is a workbench with proof, not a chat window: on your instruction it plans, uses tools and acts on your files, mail, calendar and connected software — and every outside connection is counted and reportable. Cloud assistants run on the vendor's machines. OYA runs on yours and counts every outside connection.
Does it work with my Ownware apps?
Yes — every Ownware tool ships /mcp with a bearer key from its API-settings page. Add it as a piece (or in ~/.oya/config.json) and you connect it; the product's own roles and approval guards apply to every call. The recipe is in the docs, and it was verified against live products.
Will it get updates?
The core is frozen on purpose: security fixes and honest maintenance, no feature churn. What you buy is what it stays. Pieces — MCP servers, plugins, skills, data packs — are where growth happens, and each says where its data goes.
If I use OYA to generate content for my own users, do EU AI rules apply to me?
That is a question for your adviser, not a claim this page makes. OYA is software you run on your own machine, with a model you choose; what it produces and where it goes is your decision. If you use it to generate content for your users, EU transparency rules may apply to you as the deployer — take that question to your adviser.
Can I use it for clients or in my company?
One purchase covers your own machines (personal) or your organization's internal use, source included; no redistribution or reselling. Reports and attestations are evidence, not certification.
Is Own Your AI a subscription?
No. You buy it once and own that copy — the Personal license covers one person on their own machines and the Business license one organization with unlimited internal seats. Nothing renews. The model you point it at is your own arrangement with whoever provides it.
What do I get that a hosted assistant does not give me?
The source, and the count. It runs on a model you choose and it counts every outside connection it makes, which is a thing you can audit rather than a promise you are given. In exchange you run it yourself: there is no service to call when your machine is the one that broke.