Raiana API Reference

OpenAI-compatible API for medical device regulatory intelligence — MDR, IVDR, FDA, and EU AI Act

The Raiana API provides programmatic access to our regulatory AI assistants: ChatMDR, ChatIVDR, ChatFDA, and ChatAIAct. Each assistant is augmented with up-to-date regulatory texts, MDCG guidance documents, FDA guidances, and applicable standards.

The API follows the OpenAI Chat Completions format, making it easy to integrate using the OpenAI Python library, any OpenAI-compatible client, or plain HTTP requests.

Base URL

All API endpoints are relative to this base URL. For example, the chat completions endpoint is available at https://api.chatmdr.eu/v1/chat/completions.

Authentication

All requests must include an API key in the Authorization header using the Bearer token scheme:

Authorization: Bearer raikey_YOUR_API_KEY_HERE

Raiana API keys use the raikey_ prefix. You can obtain an API key from your Raiana portal account.

Endpoints

Create Chat Completion

Sends a conversation to one of the Raiana regulatory assistants and returns a response grounded in regulatory source texts.

Request Body Parameters

Message Roles

Example Request

curl -X POST https://api.chatmdr.eu/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer raikey_YOUR_API_KEY_HERE" \
  -d '{
    "model": "chatmdr-fast-openai",
    "messages": [
      {
        "role": "user",
        "content": "What is a Class III medical device under MDR?"
      }
    ],
    "max_tokens": 1024
  }'

Response Format

Responses follow the OpenAI Chat Completion format:

{
  "object": "chat.completion",
  "id": "825c3790-ec66-4008-8b30-99efd968707c",
  "model": "chatmdr-fast-openai",
  "created": 1751373000,
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Under MDR, Class III devices are the highest risk class..."
      }
    }
  ],
  "usage": {
    "input_tokens": 5820,
    "output_tokens": 387,
    "total_tokens": 6207
  }
}

Response Fields

List Models

Returns a list of all available models.

curl https://api.chatmdr.eu/v1/models \
  -H "Authorization: Bearer raikey_YOUR_API_KEY_HERE"

Retrieve Model

Returns details about a specific model.

curl https://api.chatmdr.eu/v1/models/chatmdr-fast-openai \
  -H "Authorization: Bearer raikey_YOUR_API_KEY_HERE"

Available Models

Each model targets a specific regulatory domain and comes in two tiers: Fast for quick answers, and Smart for deeper analysis.

ChatMDR — EU Medical Device Regulation

Expert on Regulation (EU) 2017/745, MDCG guidance documents, and harmonised standards for medical devices in Europe.

ChatIVDR — EU In Vitro Diagnostic Regulation

Expert on Regulation (EU) 2017/746 (IVDR), MDCG guidance documents, and standards for in vitro diagnostic devices.

ChatFDA — FDA Medical Device Regulation

Expert on 21 CFR Title 21 (Parts 800–898), FDA guidance documents, and relevant standards for medical devices in the United States.

ChatAIAct — EU AI Act

Expert on the EU Artificial Intelligence Act (Regulation (EU) 2024/1689) and its implications for AI systems, including those used in or as medical devices.

Python Example

The Raiana API is compatible with the OpenAI Python library. Simply point the base_url to the Raiana API:

from openai import OpenAI

# Initialize the client with the Raiana API endpoint
client = OpenAI(
    base_url="https://api.chatmdr.eu/v1",
    api_key="raikey_YOUR_API_KEY_HERE",
)

# Ask a question about the EU Medical Device Regulation
response = client.chat.completions.create(
    model="chatmdr-smart-openai",
    messages=[
        {
            "role": "user",
            "content": "What are the requirements for a Quality Management System under MDR?",
        }
    ],
    max_tokens=1024,
)

# Print the assistant's answer
print(response.choices[0].message.content)

Multi-Turn Conversation

You can pass conversation history in the messages array to enable follow-up questions:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.chatmdr.eu/v1",
    api_key="raikey_YOUR_API_KEY_HERE",
)

# Build a multi-turn conversation
messages = [
    {"role": "user", "content": "Is a software that predicts disease progression a medical device under MDR?"},
    {"role": "assistant", "content": "Yes, software intended for diagnosis or prediction of disease is considered a medical device under MDR Article 2(1)..."},
    {"role": "user", "content": "What classification would it have?"},
]

response = client.chat.completions.create(
    model="chatmdr-smart-openai",
    messages=messages,
    max_tokens=1024,
)

print(response.choices[0].message.content)

Using a System Prompt

You can add a system (or developer) message to provide additional instructions. These are applied on top of the built-in Raiana system prompt:

response = client.chat.completions.create(
    model="chatmdr-smart-openai",
    messages=[
        {
            "role": "system",
            "content": "Always answer in German. Focus on Class IIa devices.",
        },
        {
            "role": "user",
            "content": "What are the clinical evaluation requirements?",
        },
    ],
)

Error Handling

The API returns standard HTTP status codes. Error responses include a JSON body with a message field:

# Handle errors gracefully
from openai import OpenAI, APIError

client = OpenAI(
    base_url="https://api.chatmdr.eu/v1",
    api_key="raikey_YOUR_API_KEY_HERE",
)

try:
    response = client.chat.completions.create(
        model="chatmdr-smart-openai",
        messages=[{"role": "user", "content": "What is Article 10?"}],
    )
    print(response.choices[0].message.content)
except APIError as e:
    print(f"API error: {e.status_code} — {e.message}")

Differences from the OpenAI Chat Completions API

The Raiana API implements a focused subset of the OpenAI Chat Completions API. The following OpenAI features are not supported:

Rate Limits & Billing

Usage is tracked per API key. Your token balance is deducted based on the usage field returned with each response. The input token count includes the regulatory context that Raiana injects, so it will be higher than just your message text.

Check your remaining balance and manage your API keys in the Raiana Portal.