Skip to main content

​
Introduction

GAIA AI provides a comprehensive REST API that enables you to:
  • Build AI Applications: Create intelligent agents with multiple AI providers
  • Manage Knowledge: Upload and search through documents with RAG capabilities
  • Execute Tools: Run custom tools and integrate external services
  • OpenAI Compatible: Use OpenAI-compatible chat completions endpoint
  • MCP Integration: Connect Model Context Protocol servers for extended functionality

​
Authentication

All API requests require authentication using a Bearer token in the Authorization header: 
Authorization: Bearer YOUR_API_KEY

​
Getting Your API Key

  1. Navigate to the API Keys page in your GAIA AI dashboard
  2. Click “Create New Key” button
  3. Copy your generated API key and store it securely
Create API Key - LightAPI Key Created - Light
Keep your API key secure! Never commit it to version control or share it publicly.

​
Base URL

​
Local Development

# Docker deployment
http://localhost:5679/api/v1

# Local development server
http://localhost:3000/api/v1

​
Production

For production deployments, replace localhost with your server address: 
https://your-domain.com/api/v1

​
Using HTTPS with Ngrok

To use GAIA AI with services that require HTTPS (like ChatGPT’s “Talk to a website” feature), you need to expose your local instance via ngrok:

​
Quick Setup

Setup ngrok: You can check it here ngrok
Your ngrok URL changes each time you restart it. For a permanent URL,

​
Available API Routes

AI

Manage AI providers, models, and configurations

Chat

OpenAI-compatible chat completions and streaming

Projects

Create and manage agent projects

Knowledge

Upload, search, and manage knowledge bases

MCP

Configure Model Context Protocol servers

Tools

Create and execute custom tools

Credentials

Manage API keys and provider credentials
https://images.seeklogo.com/logo-png/59/2/ollama-logo-png_seeklogo-593420.png

Ollama

Manage local Ollama models and configurations