Skip to content

Platform

Secure retrieval for AI, entirely inside your perimeter.

foundation4 is the search and retrieval layer for AI applications. You run it on your own infrastructure, next to your own models and your own database.

Request a briefingRead the docs
Your security perimeterYour applications, foundation4, your models and your PostgreSQL database all sit inside your security perimeter. Nothing connects out to a foundation4 cloud.YOUR SECURITY PERIMETERYour applicationsChat, portals, assistantsfoundation4Ingest, encrypt, searchREST APIYour contentSent through the APIYour modelsOpenAI-compatiblePostgreSQLYour databaseNo vendor cloudNo telemetryNo license server
Every part foundation4 needs runs inside your perimeter. Your content leaves only if an administrator registers a model server outside it.

How it works

Four steps, all on your infrastructure.

  1. Step 1

    Ingest

    Your application sends text to foundation4 through the REST API, with the classification and metadata for each item.

  2. Step 2

    Store

    foundation4 encrypts the stored text with a separate key for each classification, in your own PostgreSQL database.

  3. Step 3

    Search

    Your application searches by meaning and by exact words. Results are scoped to each reader by the classifications and metadata your application supplies.

  4. Step 4

    Answer

    Your application passes the results to any OpenAI-compatible model you run, to answer questions and power assistants.

Explore

What makes it fit where others cannot.

Security

Your content stays inside your perimeter, encrypted with a separate key for each classification.

Learn more

Deployment

Built for isolated networks, light enough to accredit, and works with the models you already run.

Learn more

Search and API

Search by meaning and by exact words, through a REST API ready for production integration.

Learn more