# GitHub Copilot Setup Guide

> Point a custom OpenAI-compatible endpoint at LLMoxy using BYOK in Copilot CLI or the Copilot app.

Website: [https://github.com/features/copilot](https://github.com/features/copilot)

GitHub Copilot completions use GitHub-hosted models by default. To send requests to LLMoxy, use **BYOK** (bring your own key). For the VS Code chat window, see the [VS Code Agents Window Setup](https://en.llmoxy.com/docs/code-tutorials/vscode-agents).

Before you start, get an [LLMoxy token](https://en.llmoxy.com/docs/get-started/get-token). The **auto** group is recommended. The model must support tool calling and streaming.

## Copilot CLI

```bash
export COPILOT_PROVIDER_TYPE=openai
export COPILOT_PROVIDER_BASE_URL=https://llmoxy.com/v1
export COPILOT_PROVIDER_API_KEY=your LLMoxy API Key
export COPILOT_MODEL=grok-4.7
copilot
```

| Variable | Value |
| --- | --- |
| `COPILOT_PROVIDER_BASE_URL` | `https://llmoxy.com/v1` |
| `COPILOT_PROVIDER_API_KEY` | LLMoxy token |
| `COPILOT_MODEL` | The exact ID from the console |
| `COPILOT_PROVIDER_TYPE` | `openai` (default) |

For Claude models, change it to:

```bash
export COPILOT_PROVIDER_TYPE=anthropic
export COPILOT_PROVIDER_BASE_URL=https://llmoxy.com
export COPILOT_PROVIDER_API_KEY=your LLMoxy API Key
export COPILOT_MODEL=claude-opus-5-5
```

## Copilot app

1. Open the GitHub Copilot app.
2. Go to **Settings → Model Providers → Add provider**.
3. Choose OpenAI-compatible (or any compatible HTTP endpoint).
4. Enter `https://llmoxy.com/v1` as the Base URL, paste your API Key, and save.
5. Pick the model you just added in the session's model selector.

This is the markdown version of https://en.llmoxy.com/docs/code-tutorials/github-copilot.
