Getting Started

Authentication

All API requests require authentication using an API Token. When you sign up for MCP GNNAIS services, you'll receive an API Token that grants you access to the API.

Authorization Header Example:

Authorization: Bearer YOUR_API_TOKEN

Environment Setup

We provide a Postman Collection for easy API testing and integration. Follow these steps to set up your environment:

  1. Download the MCP GNNAIS API Postman Collection (Download Now)
  2. Import the collection into Postman
  3. Set up your environment variables:
    Variable Description
    baseUrl https://mcp.gnnais.com/api
    API_KEY Your personal API key received after signing up

Making Your First Request

Let's make a simple request to the Cognition API using OpenAI integration:

Request Example:

{
  "generative_source": "open-ai",
  "model": "gpt-4.1-nano-2025-04-14",
  "instruction": "You are a helpful AI assistant that provides accurate and concise information.",
  "prompt": "What is artificial intelligence?"
}

API Reference

Cognition API

The Cognition API provides endpoints for text-based AI processing using various OpenAI models.

Text Completion (OpenAI)

POST

Generate text completions based on a prompt using various OpenAI models.

/cognition/make-completion

Request Body
{
  "generative_source": "open-ai",
  "model": "gpt-4.1-nano-2025-04-14",
  "instruction": "You are a helpful AI assistant that provides accurate and concise information.",
  "prompt": "whats is www.mcp.gnnais.com?"
}
Parameters
Parameter Type Description
generative_source string The AI provider (use "open-ai" for OpenAI models)
model string The specific model to use (e.g., "gpt-4.1-nano-2025-04-14")
instruction string System message to guide the model's behavior
prompt string The user's question or input text
Response
{
  "success": true,
  "http_code": 200,
  "completion": "MCP GNNAIS (www.mcp.gnnais.com) is a next-generation AI integration platform that provides multicognitive processing capabilities. It integrates various generative AI models to process and solve complex user requests through a REST API interface. The platform allows users to access different AI models, including OpenAI models, through a secure authentication system.",
  "usage": {
    "prompt_tokens": 29,
    "completion_tokens": 178,
    "total_tokens": 207
  },
  "generative_source": "open-ai",
  "model": "gpt-4.1-nano-2025-04-14"
}

Vision API

The Vision API provides endpoints for image-based AI processing using OpenAI models.

Image Interpretation (OpenAI)

POST

Analyze and interpret images using OpenAI's Vision model.

/vision/image-interpreter

Request Body
{
  "generative_source": "open-ai",
  "model": "gpt-4.1-mini",
  "instruction": "You are a helpful AI assistant that provides accurate and concise information.",
  "prompt": "What objects can you see in this image? Provide a detailed description.",
  "image": "https://mcp.gnnais.com/sx/assets/mcp-icon.png"
}
Parameters
Parameter Type Description
generative_source string The AI provider (use "open-ai" for OpenAI models)
model string The specific model to use (e.g., "gpt-4.1-mini")
instruction string System message to guide the model's behavior
prompt string The user's question about the image
image string URL of the image to analyze
Response
{
  "success": true,
  "http_code": 200,
  "analysis": "In this image, I can see the MCP GNNAIS logo. It appears to be a stylized icon with a modern, minimalist design that likely represents the company's focus on AI and technology. The icon uses a purple or blue gradient color scheme, which is common in tech and AI branding to convey innovation and creativity. The design seems to have abstract elements that might symbolize neural networks or AI processing capabilities.",
  "usage": {
    "prompt_tokens": 53,
    "completion_tokens": 163,
    "total_tokens": 216
  },
  "generative_source": "open-ai",
  "model": "gpt-4.1-mini"
}

Available OpenAI Models

MCP GNNAIS API supports all OpenAI models. You must specify which model you want to use in your request. Below are some of the popular models available:

Model Description
gpt-4.1 Latest OpenAI model with advanced capabilities
gpt-4.1-nano Fastest GPT-4.1 model variant
gpt-4.1-mini Compact GPT-4.1 model with good performance
gpt-4o GPT-4 Omni model with text and image capabilities
gpt-4o-mini Smaller version of GPT-4o with lower cost
gpt-4 OpenAI's high-capacity text model
gpt-3.5-turbo Balanced model for most text generation tasks

When making API requests, simply specify your preferred model in the "model" parameter. Our system will automatically route your request to the appropriate OpenAI model.

Rate Limits & Error Handling

Rate Limits

Rate limits vary based on your subscription plan. If you exceed your rate limit, the API will return a 429 Too Many Requests response.

Error Handling

Our API uses standard HTTP response codes. Here are some common error codes:

Status Code Description
200 Success
400 Bad Request - The request was malformed
401 Unauthorized - Authentication failed
403 Forbidden - You don't have permission
429 Too Many Requests - Rate limit exceeded
500 Internal Server Error - Something went wrong on our end