Cortexa · private AI
AI that never leaves your building.
A private chat app for your team and an OpenAI-compatible API for your apps, running on servers you own. No document goes to a public AI service.
- Runs on your network
- OpenAI-compatible API
- The AI inside CapturePoint 6
Your team wants AI. Your documents can’t go to a public cloud.
Public AI tools see everything you paste in.
Cortexa runs on your hardware. Prompts and answers stay inside.
Developers need an AI API they can trust.
An OpenAI-compatible API, with a key and usage for every app.
Staff use chat tools IT never approved.
A private chat app that signs in with your own directory.
For developers
Change one address. Keep your code.
Apps written for the OpenAI API point at your Cortexa server instead, with a Cortexa key. Same endpoints, and your data stays inside.
- Works with OpenAI-compatible SDKs and tools
- One API key per application
- Usage you can see by key and by person
client = OpenAI(
base_url="https://api.openai.com/v1",
base_url="https://cortexa.your-network/v1",
api_key=CORTEXA_KEY,
)
- POST
/v1/chat/completionsChat with a model - POST
/v1/completionsText completion - POST
/v1/embeddingsVectors for search - POST
/v1/ocr/pdfRead a PDF - POST
/v1/ocr/imageRead an image
For everyone else
A private assistant your IT team approves.
Your people get a chat app that signs in with your directory and talks only to your own server.
Private chat for your team
A familiar chat app that talks only to your Cortexa server, with history people can search and share.
Your sign-in, your rules
Single sign-on with your directory, including Active Directory, plus roles and usage limits per person.
Reads documents
Turn PDFs and images into text on your own hardware, ready for search, filing or your own tools.
Already at work
The AI already reading your documents.
Cortexa reads every page in CapturePoint 6 and Paige: it splits stacks, works out each document type, and pulls out fields and full line-item tables, learning as it goes.
Sized and quoted for your setup.
Tell us how many people will use it, what you want to connect and what it should read. We'll recommend the right Cortexa server and quote it.
Cortexa questions
What is Cortexa?
Cortexa is Ademero’s AI. It already reads documents inside CapturePoint 6 and Paige. Cortexa is also available as a private AI server, so your own team and applications can use AI on hardware you own.
What can my applications do with a Cortexa server?
Cortexa offers OpenAI-compatible chat, completion and embedding endpoints, plus OCR for PDFs and images. Applications written for the OpenAI API can use Cortexa by changing the address they call and using a Cortexa API key.
Does my data leave my network?
No. A Cortexa server runs your models on your own hardware, so prompts, documents and answers stay inside your network.
How is Cortexa priced?
Cortexa is quoted for your servers and workload. Tell us about your setup on the pricing page and we will send pricing that fits, or book a free live demo and we will walk you through it.
Free live demo
See private AI work on your own documents.
Book a free demo and we will show you around, answer your questions and run your real paperwork through it. No cost, no pressure.
- A live tour of the products that fit your work
- Your own documents, set up and shown working
- Your workflow and process, mapped with you
- Straight answers from people who build it
Sell Cortexa to your own customers? See the partner program