Straight answers.
If your question is not here, ask us directly.
What does NectarBox do?
NectarBox is one door between your apps and AI models that run in house. An app sends a standard chat request to one address with its own key. NectarBox checks who is calling, records the call (who, which model, how many tokens) and passes it to a local model. The answer goes back to the app.
Who is it for?
Teams that want to use AI on other people’s data without sending it to an outside AI company: agencies, law firms, accountants, and in-house teams that run several AI-driven apps and want one place for keys and usage. Very small teams that use AI a little will probably be better served by a plain subscription.
Does our text go to a cloud AI company?
Not in version 1. Only local models are set up, no cloud provider is configured, and so no prompt or answer is sent to one.
Do we have to rewrite our apps?
No. An app that speaks the standard chat API format most AI apps already use only needs a new address and a new key. Answers in a fixed data shape are supported. Which model answers is set in NectarBox, not in the app.
Which models does it use, and does it work with other AI services?
Local open models, chosen by setting, so apps need no code change. Which models fit depends on your hardware and needs, and we decide that with you. NectarBox is the door your apps call; it does not replace or resell outside AI services.
Who can see our prompts and answers?
By default nobody, because they are not stored. Only the facts of each call are kept: who, which model, how many tokens and when. If your organisation and an app’s key both switch the text log on, the person who owns that key can see the logged text. Our staff manage accounts and seats but never read prompts, answers or logged text.
Can we keep one client’s work apart from another’s?
Yes, that is what client walls are for. Every record is fenced by person and by place (organisation, client) in the database itself, so one client’s requests, records and cost never reach another client. This is built and in testing.
How do our apps and people get access?
Each person gets a seat and a welcome link and makes a passkey, so there are no passwords to leak. They then make a key for each app, script or bot, and see it once. A key can be suspended, revoked or handed to a colleague without touching other apps, and there are no shared keys, so every call traces to one person. You can see usage by person, app and organisation.
What do we need to run it?
A server that can run small open AI models. We size it with you. On our test box small models run on an ordinary server processor and an answer takes from a few seconds to about a minute; a server with a dedicated graphics processor would be faster and allow larger models. It can run in IT Labs India’s cloud or on your premises.
How does a pilot work?
You tell us which apps should use AI and what data they handle. We set up the box with your organisation. People get a seat and a welcome link and make a passkey. They make a key for each app and point the app at the new address. Then we review usage together and you decide what to keep. We agree the scope and length with you at the start.
Is NectarBox ready to use?
It is built and in testing. Version 1 runs on our test box, and nothing is marked “Available now” yet. Anything that does not pass testing will be removed. It has not been independently audited, and we claim no compliance certification. The features page shows the stage of each feature.
How much does it cost?
Cost is scoped per pilot, so please talk to us and we will go through it with you.
Still have a question?
Ask us, or book a pilot and bring the hardest ones.