> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tikway.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Send your first request to the Tikway API.

## 1. Create an API key

Visit [Tikway](https://tikway.ai), sign in, and create an API key in the console. Keep the key on your server. Do not put it in browser code, a public repository, or a screenshot.

## 2. Send a request

Replace `YOUR_API_KEY` with your key and choose a model available to your account:

```bash theme={null}
curl https://api.tikway.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.6-terra",
    "messages": [
      {"role": "user", "content": "Which animal would you interview first, and why?"}
    ]
  }'
```

## 3. Read the response

For Chat Completions, the generated text is usually in `choices[0].message.content`. You can change `model` to use another compatible model and pass earlier turns in `messages`.

<Note>
  Keep API keys in server-side environment variables. See [Gateway and authentication](/en/develop-guide/gateway-auth) for the OpenAI, Anthropic, and Gemini headers.
</Note>

## Next steps

<CardGroup cols={2}>
  <Card title="Request protocols" icon="code" href="/en/request-protocol/open-ai">
    Choose an OpenAI, Anthropic, or Gemini request format.
  </Card>

  <Card title="Errors and retries" icon="circle-exclamation" href="/en/develop-guide/status-code-response">
    Learn how to handle status codes, rate limits, and stream errors.
  </Card>
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.