Skip to content

Optima Documentation

Optima is a private enterprise AI platform that runs on-premises or within infrastructure you control. It lets your organisation use AI without sending sensitive information to external AI providers. It combines locally hosted AI models with your internal knowledge and live business data, using human-coded and tested Connectors to provide specific, predefined capabilities.

You can create purpose-specific AI Assistants, choose which information and tools each Assistant can access, and control who can use them. This brings AI into everyday business processes while keeping your data and systems under your control.

The Three Components

Optima is made up of three components that work together.

Optima Gateway is the central hub. It handles sign-in, request routing, users, Assistants, Knowledge Bases and the administration pages. Administrators configure everything — users, API Keys, connected Nodes and AI models — through the Gateway.

Optima Node is the inference engine. It discovers AI model files on its host machine, manages the processes that run them, and answers the requests the Gateway sends it. You can connect several Nodes to one Gateway.

Optima Chat is the app people use to talk to the AI. It runs in a web browser, on Windows, macOS and Linux, and on iPhone, iPad and Android.

Where To Start

Use the table below to find the part of the documentation you need.

I want to… Go to…
Install Optima for the first time Getting Started
Follow a guide for a common job How-To Guides
Understand the architecture Introduction
Configure the Gateway Gateway
Set up a Node Node
Start chatting Optima Chat
Install the chat app Getting the App
Connect to my services and apps Connectors and MCP
Connect via the API API