Enterprise RAG Platform

Secure, grounded AI powered by your enterprise knowledge: databases, SOPs, APIs, document repositories and social networks.

Start free 14 days · 320 questions · No card

Paste a few lines about your business and ask your first question in minutes.

Read the security brief
Assistant Scripted preview
How many orders shipped late last month?
14 of 212 orders shipped after their promised date — 6.6%.
Which customers were affected?
Three customers account for 9 of them:
CustomerLateAvg days
Northwind Foods41.8
Adriatic Retail32.3
Vilna Grocers21.5
What do we owe them for a late order?
Your Late deliveries procedure: call the customer within one working day, and credit 5% of the order when it arrives more than two days late.1
Where is #10471 now?
Out for delivery with the carrier, due today by 18:00.
Did we tell customers about the delays?
Yes. Your Facebook page posted on 29 August that orders from the central warehouse could arrive a day late until 5 September, and your website’s delivery page says the same.
Will we lose any of those customers?
Your data can't answer that — it records what happened, not what a customer will do next, and I don't guess. I can show repeat-order rates for those three accounts instead.
Features

One platform under your software: connect, ground, govern, deliver

An enterprise knowledge AI platform, powered by retrieval-augmented generation (RAG): it answers from what it finds in your own information.

Connect

  • Answers from your records

    Ask a question the way you would ask a colleague, and get the real figure from your live data.

  • Answers from your documents

    Upload manuals and policies, or point it at your website; on Growth and Scale, drives and social pages too.

  • Answers about your products

    Give it your shop’s product feed, and visitors get the products with pictures and prices. For shops

Ground

  • Says so when it does not know

    No guessing. When your material cannot answer, it tells you.

  • Checked after every change

    The questions you trust are asked again whenever something could move an answer. How it stays right

Govern

  • Never changes anything

    It can look but not edit. It has no way to change or delete a record.

  • Each customer sees only their own

    If you serve many customers, each one only ever gets their own information.

Deliver

  • Looks like yours

    In your app, a hosted page or your public site, with your name, colours and greeting. No code to change.

The platform in full

Everything it can do

In use

Our first two customers, and the jobs they gave it

A school runs it for its own people; a government institution is testing it on its public website. Neither is named here until they choose to be.

  • A local school

    In production

    Read more

    The people the school has invited ask on the page we host, from a computer or in the app on their phone. It answers from the school’s own records, its written procedures and the latest posts on its social pages — and when someone asks for one of the school’s documents, it fills in the school’s own template, shows them every value, and emails the finished file once they say yes.

    • Live records
    • Written procedures
    • Social pages
    • Documents from its templates
    • Hosted page and app

  • A government institution

    In testing

    Read more

    On its public website it answers anyone, with no sign-in, from what the institution has published and the latest posts on its social pages, and it calls the institution’s own API for the answers no page holds.

    • Public website
    • Its own API
    • Social pages
How it works

Up and running in three steps

Most teams are asking their first question the same day.

1 In your dashboard

Connect what it should know

Link your database, or upload your documents. There is nothing to install.

2 In your dashboard

Choose what it may see

You approve exactly which information it can read. Nothing is looked at until you do.

3 In your app

Add one line, and ask

Your team asks right where they already work. No change to your app? Share a page we host instead.

How it works, in detail

Data that must stay inside your own network? See the Gateway; for a product with many customers, read the ISV guide.

Applications

Applications built on the platform

Five ways to put one set of sources and rules to work.

  • The assistant

    In your app, a hosted page, the phone or your public site. Where it runs

  • The API and integrations

    Your software, Slack or Teams ask over the API. Integrations

  • Documents from your templates

    Your Word templates, filled and emailed on a yes. Templates

  • Product recommendations

    Products as cards. It recommends; your shop sells. For shops

  • A knowledge base on any subject

    Any subject your documents cover. Knowledge base

Security

Safe by design, and said plainly

Four promises, and the page that shows how each one is kept.

Your data

Read-only

It can look at your data. It cannot change it, and nothing it runs can write to your database.

Your data

Kept apart

Each of your customers sees only their own information, never anyone else’s.

Your data

Not stored

No question, no answer, no row of yours is written down here — except what a visitor chooses to send you.

Your data

Never used to train AI

We train no models, and your data is not used to improve the service for anyone else.

Read the security brief Print it to PDF and send it to whoever has to sign this off.
Pricing

Flat pricing, no token markup

Three plans, and one choice inside each: bring your own AI key, or run on ours. Pay yearly and get 2 months free. Prices in USD, excluding VAT.

What each key choice changes

What you pay us is flat and does not move when your usage does. Nothing here is sold by the seat, so the bill does not grow when the team does.

Your own key: you pay that provider directly, under your own agreement with them — included balances are higher because the model cost is not ours.
Sonela's key: nothing to set up. Your rows reach the provider we run, under our agreement, on a model we pick for cost — that cost is ours, which is why those prices are higher and the included balances lower.

Trial

Always on Sonela's AI key

Free

14 days, no card. After 14 days or 320 credits the workspace pauses; everything you configured survives and resumes when you choose a plan.

  • 1 connected data source
  • 320 question credits
  • Runs on Sonela's AI key — nothing for you to supply
Read more
  • Full security model — nothing held back
  • Email support
  • 1 website read and kept current
  • The hosted page for up to 3 people
  • 1 assistant key
Start free trial
Most popular

Growth

$70 /month

For a live assistant inside one product or workspace.

  • 3 connected data sources
  • 8,000 question credits / month — a question costs one, or two if it reads your records
  • Your AI key, billed to you by your provider
Read more
  • Custom branding, your own greeting
  • Voice input for your users — a spoken question counts as three
  • An API for your own software, on up to 5 secret keys
  • Up to 3 websites read and kept current
  • Up to 3 social pages
  • Up to 25 files each from Google Drive and SharePoint
  • Documents filled in from your Word templates
  • The hosted page for up to 25 people
  • Up to 5 assistant keys
  • Usage insights per customer
  • Business-hours support
Start free trial

Scale

$259 /month

For vendors embedding the assistant across a customer base.

  • Unlimited connected data sources
  • 40,000 question credits / month — a question costs one, or two if it reads your records
  • Your AI key, billed to you by your provider
Read more
  • Multi-tenant embedding with signed identity
  • Voice input for your users and your public assistant — a spoken question counts as three
  • White label: your logo, and no mention of Sonela
  • An API for your own software, on as many secret keys as you need
  • Websites read and kept current, as many as you need
  • Social pages, as many as you add
  • Files from Google Drive and SharePoint, as many as you share
  • Documents filled in from your Word templates
  • The hosted page for everyone you invite
  • As many assistant keys as you need
  • Priority support, onboarding help
Start free trial

Need more, or something on-premise? Tell us what you need.

Compare

Sonela, side by side

Weighing it against a database query tool, a website chatbot or a general AI assistant?

See the comparisons

Questions

The things people ask first

Anything else, write to us. A person reads it, and a person replies.

What do we need to start a trial?

An email address. No card, and no AI key — a trial runs on Sonela's key, on a model we pick, under our agreement with the provider. It lasts 14 days or 320 question credits, whichever comes first.

You do not need your data connected on day one. Start by giving the assistant your own written material — procedures, product notes, policies — and it answers from that, and says plainly that it cannot see live records yet. Bring your operational data in through the gateway when you are ready: it is the same workspace, and everything you set up stays set up.

We can’t add a script tag to our app. Can we still use it?

Yes. We host the assistant on a page of our own instead, and you invite people to it by email address — they sign in with a short code, so there are no accounts or passwords for you to manage. The gateway and the schema you approved are unchanged.

Be aware of what it costs: on our page nobody can send the screen they are stuck on, because our page is not your application. That is the one capability the embedded assistant has and this one cannot. It is the quicker way in rather than the better one, and you can move to the script tag later without setting anything up twice.

Can the assistant change or delete our data?

No. It can only look. It cannot change, add or delete anything, and that is enforced three separate ways.

Our data is on our own network. Does that work?

Yes — that is exactly what the Sonela Gateway is for. It is one small service you run on one server inside your own network, next to the data it reads. It reaches out to us and asks whether there is work; we do not dial into it.

Everything that matters happens on your side of the firewall: the gateway checks the query, applies your tenant filter, runs it read-only and sends back the rows. Nothing lands on anyone's desk, and there is nothing for your team to install on staff machines.

Do you train models on our data?

No. We train nothing, and we don't use your data to improve the service for anyone else.

Your rows go to the AI provider that answers the question and no further: the one you chose under your own terms if you brought a key, or one of the providers we run under ours if you did not. We hold accounts with more than one so an outage cannot take your assistant down, and a question reaches a second only when the first cannot answer it.

What happens when it doesn't know the answer?

It says so. Sonela answers from your own material — a query against your live operational data, or the procedures you gave the workspace — so when neither can support an answer it reports that instead of guessing: a wrong number delivered confidently is worse than no number at all.

All questions and answers

See it answer a question from your own data

Open a workspace yourself and be asking it questions in minutes — or write to us first and we will walk you through it.

Talk to us

14 days, 320 question credits, no card. It won't guess on your data. No AI key to supply — trials run on Sonela's key.