Adding AI to Visual Studio Code using GitHub Copilot and Generative APIs

Learn how to integrate AI-powered code models into Visual Studio Code (VS Code) using GitHub Copilot and Scaleway Generative APIs.

Before you start

To complete the actions presented below, you must have:

• A Scaleway account logged into the console
• Owner status or IAM permissions allowing you to perform actions in the intended Organization
• A valid API key for API authentication
• Installed Visual Studio Code on your local machine
• A GitHub account with an active Copilot subscription (free or premium)

Access GitHub Copilot in Visual Studio Code

GitHub Copilot is included by default in any up-to-date Visual Studio Code installation. You do not need to install any extension from the Marketplace.

1. Open Visual Studio Code.
2. Sign in with the GitHub account linked to your Copilot subscription.

For more details, visit the Set up Github Copilot in VS Code page from the Visual Studio Code official documentation.

Add your custom AI model to GitHub Copilot in Visual Studio Code

1. Open GitHub Copilot chat view in Visual Studio Code by selecting View > Chat in the top menu. By default, the chat view is located on the right panel of Visual Studio Code.
2. At the bottom of the chat view, click the current model name (Auto by default) to open the model dropdown menu, then click Manage Models... to open the Language Models window.
3. In the Language Models window, click Add Models to open the dropdown, then select Custom Endpoint.
4. Enter the requested information:
   • Group Name: For this example, deepseek-v4-flash-0731
   • API Key: Paste your valid API key
   • API Type: Select Chat Completions
5. The chatLanguageModels.json file opens in a new window. Edit the following fields:
   • id: deepseek-v4-flash-0731
   • name: deepseek-v4-flash-0731
   • url: https://api.scaleway.ai/v1/chat/completions
   • vision: false
6. Save and close the chatLanguageModels.json file.
7. At the bottom of the Chat view, click the current model name (Auto by default) to open the model dropdown menu, then click Other Models and select deepseek-v4-flash-0731 to start using your newly integrated model.

Going further

You can add more parameters, more models and organize them in different groups by editing the file chatLanguageModels.json.