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

# Continue

> How to use Nebius Token Factory with Continue code assistant

You can add Nebius Token Factory as a provider in [Continue](https://www.continue.dev/) code assistant within your VS Code or JetBrains IDE. It allows you to use the [models](https://docs.tokenfactory.nebius.com/ai-models-inference/overview#available-models) available in Nebius Token Factory within your IDE. The assistant can explain the code, rewrite it and provide autocomplete suggestions.

## **Prerequisites**

[Create an API key](https://tokenfactory.nebius.com/project/api-keys) for authentication.

## **Steps**

### **Install Continue**

<Tabs>
  <Tab title="JetBrains IDE">
    1. Open your JetBrains IDE and go to **Settings**.
    2. In the sidebar, select **Plugins** and search for "Continue" in the marketplace.
    3. Click **Install** next to the Continue extension.
    4. Restart the IDE.

    After installation, a Continue logo appears in your IDE sidebar. Click the logo to set up the code assistant and start using it.

    ### **Add models hosted on Nebius Token Factory**

    1. Click the Continue logo.
    2. In the chat window, open the drop-down model list at the bottom and select **Add Chat model**.
    3. In the **Provider** drop-down list, select **Nebius Token Factory**.
    4. In the **Model** drop-down list, select a model.
    5. Enter your **API key** to authorize requests.

    Now you can chat with this model or ask it to rewrite your code in the **Edit** mode. See more in the [Continue documentation](https://docs.continue.dev/getting-started/overview).

    ### **Set up a model for autocomplete**

    Code autocomplete doesn't use the model you set up for **Chat**. It supports specific [models for working with code](https://docs.tokenfactory.nebius.com/ai-models-inference/overview#available-models). You need to set the autocomplete model separately in the Continue configuration file:

    1. Click the Continue logo.
    2. In the Continue sidebar, click the gear icon, and then click **Open configuration file**.
    3. In the `config.json` file, edit the contents of the `tabAutocompleteModel` key:

       * Specify `nebius` as `provider`.
       * Enter a [model name](https://docs.tokenfactory.nebius.com/ai-models-inference/overview#available-models) for `model` and `title`.
       * Enter your [API key](https://docs.tokenfactory.nebius.com/integrations/coding/continue#prerequisites).

       For example, to use the [Qwen/Qwen2.5-Coder-32B](https://huggingface.co/Qwen/Qwen2.5-Coder-32B) model for autocomplete, enter the following:

       ```
       "tabAutocompleteModel": {
         "title": "Qwen 2.5 Coder 32b",
         "provider": "nebius",
         "model": "qwen-coder2.5-32b",
         "apiKey": "<Nebius API Key>"
       },

       ```

    See more about autocomplete in the [Continue documentation](https://docs.continue.dev/autocomplete/model-setup).
  </Tab>

  <Tab title="VS Code">
    1. Go to the [Continue extension page on Visual Studio Marketplace](https://marketplace.visualstudio.com/items?itemName=Continue.continue) and click **Install**.
    2. When you are redirected to VS Code, click **Install** again.\
       After installation, a Continue logo appears in your IDE sidebar. Click the logo to set up the code assistant and start using it.

    After installation, a Continue logo appears in your IDE sidebar. Click the logo to set up the code assistant and start using it.

    ### **Add models hosted on Nebius Token Factory**

    1. Click the Continue logo.
    2. In the chat window, open the drop-down model list at the bottom and select **Add Chat model**.
    3. In the **Provider** drop-down list, select **Nebius Token Factory**.
    4. In the **Model** drop-down list, select a model.
    5. Enter your **API key** to authorize requests.

    Now you can chat with this model or ask it to rewrite your code in the **Edit** mode. See more in the [Continue documentation](https://docs.continue.dev/getting-started/overview).

    ### **Set up a model for autocomplete**

    Code autocomplete doesn't use the model you set up for **Chat**. It supports specific [models for working with code](https://docs.tokenfactory.nebius.com/ai-models-inference/overview#available-models). You need to set the autocomplete model separately in the Continue configuration file:

    1. Click the Continue logo.
    2. In the Continue sidebar, click the gear icon, and then click **Open configuration file**.
    3. In the `config.json` file, edit the contents of the `tabAutocompleteModel` key:

       * Specify `nebius` as `provider`.
       * Enter a [model name](https://docs.tokenfactory.nebius.com/ai-models-inference/overview#available-models) for `model` and `title`.
       * Enter your [API key](https://docs.tokenfactory.nebius.com/integrations/continue#prerequisites).

       For example, to use the [Qwen/Qwen2.5-Coder-32B](https://huggingface.co/Qwen/Qwen2.5-Coder-32B) model for autocomplete, enter the following:

       ```
       "tabAutocompleteModel": {
         "title": "Qwen 2.5 Coder 32b",
         "provider": "nebius",
         "model": "qwen-coder2.5-32b",
         "apiKey": "<Nebius API Key>"
       },

       ```

    See more about autocomplete in the [Continue documentation](https://docs.continue.dev/autocomplete/model-setup).
  </Tab>
</Tabs>