Academic AI is a project involving more than 20 Austrian universities, technically and organizationally led by ACOmarket. It provides access to AI chat models on an isolated Microsoft Azure instance — meaning inputs and uploaded documents never leave the “Azure Secure Landing Zone.”

Go to mul.academic-ai.at, select “Montanuniversität Leoben,” and log in with your MUOnline credentials.

Usage is capped at €20 per person per month, resetting at the start of each month with no rollover. A token is the smallest unit of data processed by the AI — uploading and analyzing a document consumes more tokens than a typed text input.

• GPT-4o
• GPT-5
• GPT-5 mini
• GPT-5 nano
• Mistral Large 3
• o3
• Google Gemini (coming soon)
• Claude (coming soon)

Data is not shared with OpenAI or third parties. Your data is automatically deleted 90 days after your last login. Chat history can also be deleted manually at any time.

Users can create custom chatbots connected to their own documents. Collaborators can be invited with User, Contributor, or Admin roles.

The platform exposes an API for developers who want to integrate Academic AI into their own applications. It supports LLM chat completions, Tailored AI management, and Knowledge Base management for Retrieval-Augmented Generation (RAG). Authentication uses X-Client-ID and X-Client-Secret headers.

API access tokens can be requested by e-mail at academic-AI@unileoben.ac.at.

Example – Chat request (Python)

import requests

headers = {
    "X-Client-ID": "your-api-key",
    "X-Client-Secret": "your-api-secret",
    "Content-Type": "application/json",
}
payload = {
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Your question here"}],
    "temperature": 0.7,
    "maxTokens": 1000,
}
response = requests.post(
    "https://api.example.com/api/v1/llm/chat",
    headers=headers,
    json=payload
)
print(response.json()["data"]["content"])

Example – Chat request (JavaScript)

const res = await fetch("https://api.example.com/api/v1/llm/chat", {
  method: "POST",
  headers: {
    "X-Client-ID": "your-api-key",
    "X-Client-Secret": "your-api-secret",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "gpt-4o",
    messages: [{ role: "user", content: "Your question here" }],
    temperature: 0.7,
    maxTokens: 1000,
  }),
});
const data = await res.json();
console.log(data.data.content);

Full API documentation is available upon request.