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

# JanitorAI

> Connect JanitorAI to Routeway using the proxy configuration.

JanitorAI is a web-based character AI platform. You can route your JanitorAI conversations through Routeway by adding a proxy configuration.

<Note>
  Requires a valid Routeway API key. Get yours from the [dashboard](https://routeway.ai/dashboard/api-keys).
</Note>

***

## Setup

<Steps>
  <Step title="Open API Settings">
    In JanitorAI, go to **API Settings**.
  </Step>

  <Step title="Select Proxy">
    Choose **Proxy** as the provider from the list.
  </Step>

  <Step title="Add a proxy configuration">
    Click **+ Add** under Proxy Configurations and fill in:

    | Field                  | Value                                                                            |
    | ---------------------- | -------------------------------------------------------------------------------- |
    | **Configuration Name** | Any label, e.g. `Routeway`                                                       |
    | **Model Name**         | A model ID from [routeway.ai/models](https://routeway.ai/models), e.g. `gpt-4.1` |
    | **Proxy URL**          | `https://api.routeway.ai/v1/chat/completions`                                    |
    | **API Key**            | Your Routeway API key                                                            |
  </Step>

  <Step title="Activate the configuration">
    Click **Apply**, then make sure the Routeway proxy configuration is marked **Active**.
  </Step>

  <Step title="Start chatting">
    JanitorAI will now route your conversations through Routeway. Pick a character and begin.
  </Step>
</Steps>

***

## Troubleshooting

<AccordionGroup>
  <Accordion title="PROXY ERROR">
    The most common causes:

    * You selected a paid model but your Routeway account has no balance. Add funds from the [dashboard](https://routeway.ai/dashboard).
    * The proxy URL is incorrect. It must be exactly `https://api.routeway.ai/v1/chat/completions`.
    * Your API key has extra spaces — re-paste it carefully.
  </Accordion>

  <Accordion title="Network error">
    Try these steps:

    1. Re-enter your API key
    2. Refresh the page
    3. Restart your browser
    4. Try a different browser if the issue persists
  </Accordion>

  <Accordion title="Model not responding">
    Verify the model ID matches exactly. Browse available models at [routeway.ai/models](https://routeway.ai/models) or use the [models endpoint](https://api.routeway.ai/v1/models).
  </Accordion>

  <Accordion title="&#x22;Unable to process your request right now. (unk)&#x22;">
    This error is **not** caused by Routeway. It appears when JanitorAI's own backend fails internally before your message ever reaches your configured proxy. Your request never makes it to Routeway.

    <Frame>
      <img src="https://mintcdn.com/clashai-de6ad0cc/2_Y3YcKnIfODaDl3/images/Screenshot_20260610_051723_Chrome.jpg?fit=max&auto=format&n=2_Y3YcKnIfODaDl3&q=85&s=3f59187dbd3eef41fbb24e554609be6f" alt="JanitorAI unable to process request error" width="1080" height="377" data-path="images/Screenshot_20260610_051723_Chrome.jpg" />
    </Frame>

    What you can do:

    1. Wait a few minutes and try again — it usually resolves on its own
    2. Refresh the page or restart the browser
    3. Try a different character or start a new chat
    4. If it persists, JanitorAI may be experiencing downtime
  </Accordion>

  <Accordion title="Network error: You may be rate limited or experiencing connection issues">
    If you see the error message **"A network error occurred, you may be rate limited or having connection issues: Failed to fetch"**\*, the most common cause is an incorrect proxy URL.

    Verify that your proxy URL is set to:
    `https://api.routeway.ai/v1/chat/completions`

    Make sure the URL does **not** include a trailing slash (`/`).
  </Accordion>

  <Accordion title="Responses are very slow or timeout after ~90–100 seconds">
    If responses consistently take more than **90–100 seconds** and eventually fail with an error, the most common cause is that **Text streaming is disabled** in JanitorAI.

    To enable it:

    1. Open any chat
    2. Click **Customize**
    3. Find **Text streaming**
    4. Enable it

    After enabling **Text streaming**, responses should begin appearing as they're generated instead of waiting for the entire completion, preventing most 90–100 second timeouts.
  </Accordion>
</AccordionGroup>