GitHub Copilot CLI with AI Gateway

GitHub Copilot CLI is GitHub's terminal coding agent. Its bring-your-own-key mode uses environment variables, so this mode needs no config file or GitHub sign-in.

Run the Vercel CLI setup command for GitHub Copilot CLI:

npx vercel ai-gateway setup --agent copilot

The command exports COPILOT_PROVIDER_BASE_URL, COPILOT_PROVIDER_API_KEY, and COPILOT_MODEL from a managed block in your shell startup file. It doesn't write a config file.

If you can't use the Vercel CLI, configure GitHub Copilot CLI manually:

1. Create an API key

   Go to the AI Gateway section in the Vercel dashboard sidebar and click API keys to create a new API key.

2. Export the environment variables

   Copilot CLI activates its BYOK provider as soon as COPILOT_PROVIDER_BASE_URL is set. Export all three variables in your shell:

   Terminal

   export COPILOT_PROVIDER_BASE_URL="https://ai-gateway.vercel.sh/coding-agent/v1"
   export COPILOT_PROVIDER_API_KEY="your-ai-gateway-api-key"
   export COPILOT_MODEL="openai/gpt-6-astra"

   To persist them, add the same lines to your shell's startup file. COPILOT_PROVIDER_API_KEY carries the key, and COPILOT_MODEL selects the default gateway model.

3. Start Copilot CLI

   Open a new terminal so the variables are loaded, then start the agent:

   Terminal

   copilot

   GitHub sign-in is not required in this mode.

4. Switch models

   Set COPILOT_MODEL to any gateway model ID, or pass it per run:

   Terminal

   copilot --model anthropic/claude-sonnet-5

View your usage, spend, and request activity in the AI Gateway section in the Vercel dashboard sidebar. See the observability documentation for more details.

• Browse the models catalog for available models
• Configure spend limits on your API key