# Quickstart

Send your first request to Aqueduct in three steps.

## 1. Get an API key

1. Log in at [aqueduct.ai.datalab.tuwien.ac.at](https://aqueduct.ai.datalab.tuwien.ac.at/) with your TU Wien account (SSO).
2. Create a personal API key under **API Keys**.
3. Store it in an environment variable, so it never ends up in your code:

    ``` bash
    export AQUEDUCT_API_KEY="sk-..."
    ```

Students need to request access first, see [Getting Access](https://datalab.tuwien.ac.at/aiml/access/#students).

## 2. Know the endpoints

| What | URL |
| --- | --- |
| OpenAI-compatible API | `https://aqueduct.ai.datalab.tuwien.ac.at/v1` |
| MCP servers | `https://aqueduct.ai.datalab.tuwien.ac.at/mcp-servers/{name}/mcp`, see [MCP Servers](https://datalab.tuwien.ac.at/aiml/aqueduct/mcp/) |

Aqueduct is compatible with the OpenAI API, so any OpenAI client library or tool works by changing the base URL.

## 3. Send a request

=== "curl"

    ``` bash
    curl https://aqueduct.ai.datalab.tuwien.ac.at/v1/chat/completions \
      -H "Authorization: Bearer $AQUEDUCT_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "qwen-3.8-27b",
        "messages": [{"role": "user", "content": "Say this is a test"}]
      }'
    ```

=== "Python"

    ``` python
    import os
    from openai import OpenAI

    client = OpenAI(
        base_url="https://aqueduct.ai.datalab.tuwien.ac.at/v1",
        api_key=os.environ["AQUEDUCT_API_KEY"],
    )

    chat_completion = client.chat.completions.create(
        model="qwen-3.8-27b",
        messages=[{"role": "user", "content": "Say this is a test"}],
    )
    print(chat_completion.choices[0].message.content)
    ```

=== "JavaScript"

    ``` javascript
    import OpenAI from "openai";

    const client = new OpenAI({
      baseURL: "https://aqueduct.ai.datalab.tuwien.ac.at/v1",
      apiKey: process.env.AQUEDUCT_API_KEY,
    });

    const completion = await client.chat.completions.create({
      model: "qwen-3.8-27b",
      messages: [{ role: "user", content: "Say this is a test" }],
    });
    console.log(completion.choices[0].message.content);
    ```

To use a different model, replace the `model` value with any ID from [Available Models](https://datalab.tuwien.ac.at/aiml/aqueduct/models/).

## Next steps

- [Available Models](https://datalab.tuwien.ac.at/aiml/aqueduct/models/): context lengths, image input and reasoning efforts
- [Integrations](https://datalab.tuwien.ac.at/aiml/aqueduct/integrations/): set up coding agents and editors
- [Public Aqueduct documentation](https://tu-wien-datalab.github.io/aqueduct/): speech, image generation, files and batches, and more
- [Usage examples](https://cookbooks.git-pages.tuwien.ac.at/private/llm-as-a-service-cookbook/) in the LLM-as-a-Service cookbook
- [FAQ](https://datalab.tuwien.ac.at/aiml/faq/): rate limits, reasoning, timeouts