Skip to main content

​
Overview

The chat completions endpoint allows you to interact with CTGT AI models to generate responses based on conversation history. Endpoint: https://api.ctgt.ai/v1/chat/completions Method: POST

​
Authentication

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

​
Request Parameters

ParameterTypeRequiredDescription
modelstringYesThe model to use (e.g., gemini-2.5-flash-lite)
messagesarrayYesArray of message objects with role and content
personastringNoThe persona to use (default: "default")
temperaturenumberNoControls randomness (0-2, default: 1)
streambooleanNoWhether to stream the response (default: false)
userstringNoUnique identifier for the end-user

​
Available Personas

Choose from the following personas to customize the model’s behavior:
  1. default - Standard balanced responses
  2. technical_expert - Detailed technical explanations
  3. creative_writer - Creative and engaging content
  4. educator - Clear, teaching-oriented responses
  5. safety_first - Extra emphasis on safety and caution

​
Code Examples

import requests
import json

url = "https://api.ctgt.ai/v1/chat/completions"
api_key = "YOUR_API_KEY"

headers = {
    "Authorization": f"Bearer {api_key}",
    "Content-Type": "application/json"
}

payload = {
    "model": "gemini-2.5-flash-lite",
    "persona": "default", 
    "messages": [
        {
            "role": "user",
            "content": "Why is the sky blue?"
        }
    ],
    "temperature": 1,
    "stream": False, 
    "user": "user-123",
}

try:
    response = requests.post(url, headers=headers, json=payload)
    response.raise_for_status()
    print(json.dumps(response.json(), indent=2))
except Exception as e:
    print(f"Error: {e}")

​
Response Format

A successful response returns a JSON object with the following structure: 
{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "gemini-2.5-flash-lite",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "I don't have access to real-time weather data..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 12,
    "total_tokens": 21
  }
}

​
Error Handling

The API returns standard HTTP status codes:
  • 200: Successful request
  • 400: Bad request (invalid parameters)
  • 401: Unauthorized (invalid API key)
  • 429: Rate limit exceeded
  • 500: Internal server error
Example error response: 
{
  "error": {
    "message": "Invalid API key provided",
    "type": "invalid_request_error",
    "code": "invalid_api_key"
  }
}

​
Best Practices

Pro tip: Use the persona parameter to tailor responses to your specific use case without changing your prompts.
Always keep your API key secure and never expose it in client-side code. Use environment variables or secure key management systems.
  • Start with temperature: 1 for balanced responses
  • Use temperature < 1 for more focused, deterministic outputs
  • Use temperature > 1 for more creative, varied responses
  • Implement proper error handling for production applications
  • Consider using streaming for better user experience with long responses