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

# Getting Started

> Get up and running with Open Politics HQ

# Getting Started

You have two options: use our hosted instance, or run HQ yourself.

## Choose Your Setup

<Tabs>
  <Tab title="Hosted Instance">
    The fastest way to start. We handle the infrastructure - you bring your own LLM API keys.

    <Steps>
      <Step title="Register">
        Sign up at [open-politics.org](https://open-politics.org/accounts/register)
      </Step>

      <Step title="Add API Keys">
        Configure your LLM API keys (OpenAI, Anthropic, Google, or run Ollama locally)
      </Step>

      <Step title="Start Analysing">
        Upload content and create your first schema
      </Step>
    </Steps>

    <Note>
      Your account also works on the [forum](https://forum.open-politics.org) for community support.
    </Note>
  </Tab>

  <Tab title="Self-Hosted">
    For privacy, customization, or institutional requirements. Run everything on your own infrastructure.

    <Steps>
      <Step title="Prerequisites">
        Docker, Git, 32GB RAM recommended
      </Step>

      <Step title="Clone & Setup">
        ```bash theme={null}
        git clone https://github.com/open-politics/open-politics-hq.git
        cd open-politics-hq
        cp .env.example .env
        # Edit .env with your configuration
        ```
      </Step>

      <Step title="Start Services">
        ```bash theme={null}
        docker compose up
        ```
      </Step>

      <Step title="Access Platform">
        Open [http://localhost:3000](http://localhost:3000)
      </Step>
    </Steps>

    <Warning>
      **Change the defaults immediately:**

      * Username: `app_user`
      * Password: `app_user_password`
    </Warning>

    ### Deployment Options

    <CardGroup cols={3}>
      <Card title="Fully Local" icon="shield">
        Everything on your hardware. Air-gapped environments, complete data control.
      </Card>

      <Card title="Hybrid" icon="cloud">
        App locally, managed services (AWS RDS, Upstash Redis, S3) for less overhead.
      </Card>

      <Card title="Kubernetes" icon="server">
        Helm chart available at [GitHub](https://github.com/open-politics/open-politics-hq/tree/main/.deployments)
      </Card>
    </CardGroup>
  </Tab>
</Tabs>

***

## LLM Support

HQ works with multiple AI providers. Bring your own keys.

<CardGroup cols={2}>
  <Card title="Ollama" icon="shield" color="#0D9373">
    Run locally: Llama, Qwen, etc. - complete privacy
  </Card>

  <div>
    <Card title="Anthropic" icon="brain" color="#EC4899">
      Claude models - best experience with tools
    </Card>

    <Card title="OpenAI" icon="robot" color="#10A37F">
      GPT-4o and newer
    </Card>

    <Card title="Google" icon="google" color="#4285F4">
      Gemini - large context, speed, multimodal
    </Card>
  </div>
</CardGroup>

Configure keys in the web interface. Or run Ollama locally and skip the cloud entirely.