Ada Brain API Docs

Welcome to Ada! Your personal AI assistant that runs entirely on your hardware.

First time here? Check out the visual introduction 🌱 for a gentle tour!

Ready to dive in? Start here:

  1. Zero to Ada: Zero to Ada - Fastest path from nothing to working Ada (< 10 minutes)

  2. Customize: Getting Started from Scratch - Make Ada truly yours

  3. Explore: Specialist System - Extend Ada’s capabilities

Multiple setup options: Use Nix (any Linux/macOS), local Python 3.13, or Docker. Choose what works for you!

Getting Started

Hardware Setup

Core Architecture

Interfaces & Adapters

Specialist System

Operations

Development

Philosophy & Design

Research & Validation

Overview

The Brain API provides endpoints for:

  • Chat with RAG - Generate responses with context from persona, FAQ, memories, and conversation history

  • Streaming - Real-time token delivery via Server-Sent Events (SSE)

  • Memory Management - Store and retrieve long-term user/context memories

  • Health Monitoring - Check service and dependency status

  • Debug Tools - Inspect RAG system state

Key Features:

✓ Non-blocking streaming responses (see Streaming) ✓ Semantic search on memories (see Memory) ✓ Automatic conversation threading ✓ Thinking/reasoning visibility ✓ Entity-scoped context retrieval (see Data Model Reference) ✓ Conversation summarization ✓ Specialist plugin system (see Specialist System)

Quick Start

Health Check:

curl http://localhost:7000/v1/healthz

Simple Chat:

curl -X POST http://localhost:7000/v1/chat \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Hello!"}'

Streaming Chat:

curl -N -X POST http://localhost:7000/v1/chat/stream \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Hello!"}'

See Getting Started for detailed setup and configuration.

Full API Reference

For complete endpoint documentation with parameters, responses, and examples, see API Reference.