Integrate GitHub Copilot

How to use Nusabase API in GitHub Copilot workflows, including limitations, BYOK-style setup, validation, and troubleshooting.

This guide is for users who already work in VS Code with GitHub Copilot Chat and want to route part of their model traffic through Nusabase API.

The official GitHub Copilot Chat client does not always expose a general field for arbitrary OpenAI-compatible Base URLs. To use Nusabase in a Copilot Chat model picker, use a Copilot extension or BYOK workflow that supports custom providers. For a more predictable terminal setup, see GitHub Copilot CLI.

When To Use GitHub Copilot

• You mainly work in VS Code.
• You already use GitHub Copilot.
• You want to keep Copilot Chat, agent mode, tool use, and MCP workflows.
• Your Copilot extension supports custom API keys, Base URLs, and model names.

Prerequisites

• VS Code installed.
• GitHub Copilot installed and signed in.
• A valid API key: sk-....
• Network access to https://api.nusabasedigital.com.
• At least one available model from GET /v1/models.

Validate Nusabase First

Before configuring Copilot, confirm that your API key and model work.

curl https://api.nusabasedigital.com/v1/models \
  -H "Authorization: Bearer sk-your_api_key"

Then test Chat Completions:

curl https://api.nusabasedigital.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.5-flash",
    "messages": [
      {
        "role": "user",
        "content": "hello"
      }
    ],
    "stream": false
  }'

Configuration Values

If your Copilot extension supports custom models or OpenAI-compatible providers, use these values:

Setting	Recommended value	Notes
Provider type	openai-compatible or openai	The exact name depends on the extension.
Base URL	https://api.nusabasedigital.com/v1	If the extension requires a full endpoint, use https://api.nusabasedigital.com/v1/chat/completions.
API Key	sk-your_api_key	Do not paste the Base URL into the API key field.
Model	qwen3.5-flash	Example only. Use /v1/models as the source of truth.
Streaming	Enabled	Copilot Chat usually benefits from streaming output.

Common flow:

1. Open the VS Code command palette.
2. Open the extension’s Provider, Model, or BYOK configuration screen.
3. Add an OpenAI-compatible provider named Nusabase.
4. Fill in Base URL, API key, and model name.
5. Return to Copilot Chat and select the Nusabase model.

Model Strategy

Scenario	Recommendation
Daily Q&A and lightweight code explanations	Use a fast, low-cost model.
Multi-file edits and deeper reasoning	Use a stronger model with a larger context window.
Team-wide setup	Fix the provider name and default model in workspace or extension settings.

For actual model availability and pricing, see Models and Pricing.

Validation

• The Nusabase model appears in the Copilot model picker.
• A hello prompt returns a valid response.
• Errors expose Nusabase or upstream HTTP status information.

Troubleshooting

Problem	Likely cause	Fix
No Base URL field in native Copilot settings	The current client may not support arbitrary custom providers	Use a BYOK/custom-provider extension or switch to Copilot CLI.
401	Missing or invalid API key	Confirm that sk-... is in the API key field.
404 or model unavailable	The model name is not available to your account	Call /v1/models and copy the returned model ID.
Slow response	Large model, long context, or network latency	Reduce context, use a faster model, and keep streaming enabled.

Next: GitHub Copilot CLI.

References

• GitHub Copilot docs
• GitHub Copilot CLI BYOK
• GitHub Copilot CLI product page
• First Request Example
• Models and Pricing