Continue

Continue is a development tool (VS Code extension and CLI) that uses AI models to assist with coding tasks such as autocompletion, explanation, refactoring, and more. To make AI requests, Continue needs a model provider, which can be configured to use your AI Gateway.

You can view Help Guide for more details.

API Endpoint Management

1. In VS Code, install the Continue extension if not already installed.

2. Open the Continue settings:

   • Press Ctrl/Cmd + P to open the command palette.

   • Type: Continue: Open Settings.

   • Or open the extension sidebar and click the gear / settings icon for Continue.

3. Open (or create) the Continue configuration file:

   vim ~/.continue/config.json

   If the file does not exist, create it manually.

4. Paste the following example configuration and replace the apiKey with your own values:

   {
     "models": [
       {
         "title": "AI Gateway Chat Model",
         "provider": "openai",
         "model": "gpt-4o",
         "apiKey": "sk-xxxxxxxxxxxxxx",
         "apiBase": "https://gateway.theturbo.ai"
       }
     ],
     "tabAutocompleteModel": {
       "title": "AI Gateway Autocomplete",
       "provider": "openai",
       "model": "gpt-4o-mini",
       "apiKey": "sk-xxxxxxxxxxxxxx",
       "apiBase": "https://gateway.theturbo.ai"
     }
   }

Continue uses different models for different purposes:

• models Used for chat and general AI interactions, such as:

  • Asking questions

  • Explaining code

  • Refactoring

  • Generating code snippets

• tabAutocompleteModel Used specifically for code auto-completion, for example:

  • Suggestions that appear while typing

  • Pressing Tab to complete code

You can use the same model for both, or different models depending on your needs.