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

# Continue | AI Gateway

> Route Continue extension model calls through Orq.ai AI Gateway for unified observability, cost tracking, and model governance.

Route [**Continue**](https://marketplace.visualstudio.com/items?itemName=Continue.continue) model calls through the [**Orq.ai AI Gateway**](/docs/ai-gateway/get-started/quickstart). **Continue** is an open-source coding assistant extension for VS Code that supports custom OpenAI-compatible providers. Requests appear in [Traces](/docs/ai-gateway/traces) automatically.

## Prerequisites

* [Visual Studio Code](https://code.visualstudio.com/) installed
* [**Continue**](https://marketplace.visualstudio.com/items?itemName=Continue.continue) extension installed
* Active **Orq.ai** account with **AI Gateway** access
* [**Orq.ai** API key](/docs/ai-gateway/api-keys)
* Model enabled in [**AI Gateway** → **Supported Models**](/docs/ai-gateway/supported-models)

## Setup

<Steps>
  <Step title="Create or edit ~/.continue/config.yaml">
    Create `~/.continue/config.yaml` if it does not exist. If the file already has a `models` list, add the **Orq.ai** entry as a new item:

    ```yaml theme={"theme":{"light":"github-light","dark":"github-dark"}}
    models:
      - name: Orq Gateway
        provider: openai
        model: openai/gpt-5.4
        apiBase: https://api.orq.ai/v3/router
        apiKey: <ORQ_API_KEY>
    ```

    Replace `<ORQ_API_KEY>` with an [API key](/docs/ai-gateway/api-keys). To avoid storing the key in plain text, use `apiKey: $ORQ_API_KEY` and export the variable in the shell that launched VS Code. Use the `code` command to open VS Code from the terminal; VS Code launched from Dock or Spotlight does not inherit shell variables.

    Keep `openai/gpt-5.4` to route requests to that model through the **AI Gateway**, or replace it with any provider-prefixed model ID enabled in [**AI Gateway** → **Supported Models**](/docs/ai-gateway/supported-models) (e.g. `anthropic/claude-sonnet-4-6`).
  </Step>

  <Step title="Select the model in Continue">
    Open the **Continue** panel (<kbd><Icon icon="command" /></kbd>+<kbd>Shift</kbd>+<kbd>L</kbd> / <kbd>Ctrl</kbd>+<kbd>Shift</kbd>+<kbd>L</kbd>) and select **Orq Gateway** from the model picker at the bottom of the chat input.

    If **Orq Gateway** does not appear, check the top of the **Continue** panel for a config error banner.
  </Step>
</Steps>

## Configuration Reference

| Field      | Value                          | Description                                                                |
| ---------- | ------------------------------ | -------------------------------------------------------------------------- |
| `name`     | Any display label              | Name shown in the **Continue** model picker.                               |
| `provider` | `openai`                       | Tells **Continue** to use the OpenAI-compatible wire format.               |
| `model`    | `provider/model-id`            | Provider-prefixed model ID (e.g. `anthropic/claude-sonnet-4-6`).           |
| `apiBase`  | `https://api.orq.ai/v3/router` | **AI Gateway** router endpoint.                                            |
| `apiKey`   | **Orq.ai** API key             | Sent as the Bearer token. Use `$ORQ_API_KEY` to read from the environment. |

## Troubleshooting

<AccordionGroup>
  <Accordion title="Orq Gateway not appearing in the model picker">
    Confirm `~/.continue/config.yaml` was saved correctly. Open the **Continue** panel and check for a config error banner at the top. It displays YAML parse errors that prevent models from loading. Reload VS Code if the file was edited while **Continue** was already running.
  </Accordion>

  <Accordion title="Authentication error">
    Confirm the `apiKey` value is a valid **Orq.ai** API key. To avoid storing the key in plain text, set `apiKey: $ORQ_API_KEY` and export the variable in the shell that launched VS Code.
  </Accordion>

  <Accordion title="Model not found">
    The model must be enabled in [**AI Gateway** → **Supported Models**](/docs/ai-gateway/supported-models) and the `model` field must use the provider-prefixed format (e.g. `openai/gpt-5.4`, not `gpt-5.4`).
  </Accordion>

  <Accordion title="No Traces appearing in Orq.ai">
    Confirm `apiBase` is `https://api.orq.ai/v3/router` with no trailing slash. Verify **Orq Gateway** is selected in the **Continue** model picker before sending a prompt.
  </Accordion>
</AccordionGroup>

## Verification

Send a prompt in the **Continue** panel. The response appears in VS Code and a new row appears within a few seconds in [**Orq.ai** Traces](/docs/ai-gateway/traces), showing the model identifier (e.g. `openai/gpt-5.4`) and token usage.