GitHubCopilot

Use Cases

1. VS Code: Recommended Method

1. Update VS Code: Update VS Code to a stable version that includes Language Models management.

2. Open Model Management: Press Cmd/Ctrl + Shift + P to open the Command Palette, and run Chat: Manage Language Models. You can also click the gear icon in the Chat model selector.

3. Add a Custom Endpoint: Click Add Models → Custom Endpoint.

4. Enter the Basic Information: Enter MaiToken as the group name, MaiToken - Model ID as the display name, and your MaiToken Key as the API Key.

5. Select the API Type: Select Chat Completions.

6. Edit the Configuration File: VS Code will open chatLanguageModels.json. Replace <your MaiToken Key> and <Model ID> in the following example with your actual values, and then save the file.

7. Select the Model: Return to Chat and select the MaiToken model from the model selector. If the model does not appear immediately, restart VS Code.

{  "name": "MaiToken",  "vendor": "customendpoint",  "apiKey": "<your MaiToken Key>",  "apiType": "chat-completions",  "models": [    {      "id": "<Model ID>",      "name": "MaiToken - <Model ID>",      "url": "https://api.maitoken.com/v1/chat/completions",      "toolCalling": true,      "streaming": true,      "maxInputTokens": 128000,      "maxOutputTokens": 8192    }  ]}

Test the Configuration

1. Test a Basic Conversation: Enter “Reply with OK only” in the chat box. If you receive a valid response, the API Key, API address, and Model ID are configured correctly.

2. Test Project Reading: Open a test project and enter “Read the current project directory and list only the files at the first level. Do not modify any files.”

3. Test Agent Capabilities: Ask the Agent to create a file named test-maitoken.txt and write hello into it. Confirm that the file was created successfully. Do not use a real production project for testing.

2. Copilot CLI: For Terminal Users

Before starting Copilot CLI, set the following environment variables in the current terminal session.

macOS / Linux

export COPILOT_PROVIDER_BASE_URL="https://api.maitoken.com/v1"export COPILOT_PROVIDER_TYPE="openai"export COPILOT_PROVIDER_API_KEY="<your MaiToken Key>"export COPILOT_MODEL="<Model ID>" copilot

Windows PowerShell

$env:COPILOT_PROVIDER_BASE_URL="https://api.maitoken.com/v1"$env:COPILOT_PROVIDER_TYPE="openai"$env:COPILOT_PROVIDER_API_KEY="<your MaiToken Key>"$env:COPILOT_MODEL="<Model ID>" copilot

3. GitHub Copilot App

1. Open Settings: Start the GitHub Copilot App, and then open Settings → Model providers.

2. Add a Provider: Click Add provider and select OpenAI-compatible HTTP endpoint.

3. Enter the Parameters: Enter MaiToken as the display name; enter https://api.maitoken.com/v1 as the Base URL; enter your MaiToken Key as the API Key; and enter <Model ID> as the model.

4. Save and Select the Model: Save the configuration, select the model from the model selector, and then start an Agent session.

Common Issues