Skip to content

Integrating with AI Controller

AI Controller is designed to integrate seamlessly with your existing applications, frameworks, and workflows. This section covers the various integration options and provides guidance for developers.

Integration Overview

AI Controller provides several ways to integrate with your applications and tools:

  • REST API: Direct HTTP access to AI Controller endpoints
  • API Keys: Secure authentication for application access
  • Client Integration: Integration with popular AI clients and interfaces
  • Framework Integration: Support for AI development frameworks
  • Local LLM Integration: Connection to locally hosted language models

REST API Integration

The most direct integration method is through AI Controller's REST API. AI Controller provides OpenAI-compatible endpoints that make it easy to adapt existing code. This enables you to route requests through AI Controller with minimal changes to your code.

REST API Reference

API Keys for Integration

AI Controller provides a robust API key system for application authentication.

Learn more about API Keys

Local LLM Integration

AI Controller can connect to locally hosted language models running on your infrastructure, providing the same governance and monitoring capabilities as with cloud-based models while keeping data within your control.

Learn how to integrate local LLMs

Client Integrations

AI Controller seamlessly integrates with popular AI clients and interfaces to provide user-friendly access to language models while maintaining security and governance.

Explore Client Integrations

Some popular client integrations include:

  • LibreChat - An open-source chat interface supporting multiple AI models
  • Continue-Dev - A code assistance tool for developers

Framework and Library Integrations

AI Controller works with popular AI frameworks and libraries, allowing developers to build applications using familiar tools while leveraging AI Controller's security and monitoring features.

Integration Examples

import requests

# AI Controller endpoint
aic_url = "https://your-aic-server:9090/work/"

# Your AIC API key
api_key = "your-aic-api-key"

# Request payload
payload = {
    "model": "gpt-4",
    "messages": [
        {"role": "user", "content": "Hello, world!"}
    ]
}

# Send request to AIC
response = requests.post(
    aic_url,
    json=payload,
    headers={"Authorization": f"Bearer {api_key}"}
)

# Process response
if response.status_code == 200:
    print(response.json())
else:
    print(f"Error: {response.status_code}")
    print(response.text)

async function queryAIC() {
    const response = await fetch('https://your-aic-server:9090/work/', {
        method: 'POST',
        headers: {
            'Content-Type': 'application/json',
            'Authorization': 'Bearer your-aic-api-key'
        },
        body: JSON.stringify({
            model: 'gpt-4',
            messages: [
                {role: 'user', content: 'Hello, world!'}
            ]
        })
    });

    if (response.ok) {
        const data = await response.json();
        console.log(data);
    } else {
        console.error('Error:', response.status);
    }
}

curl -X POST https://your-aic-server:9090/work/ \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-aic-api-key" \
  -d '{
    "model": "gpt-4",
    "messages": [
      {"role": "user", "content": "Hello, world!"}
    ]
  }'

Enterprise Integration Patterns

AI Controller supports common enterprise integration patterns:

API Gateway Pattern

Use AI Controller as a specialized AI gateway within your existing API gateway infrastructure to route, monitor, and secure all AI model interactions.

Sidecar Pattern

Deploy AI Controller as a sidecar to applications that need AI capabilities, allowing localized routing and caching while maintaining centralized governance.

Next Steps

After reviewing the integration options, consider:

Updated: 2025-05-27