Copilot CLI

iBefore you start

GitHub Copilot CLI is installed and available as copilot.
A working Yolo Router endpoint, for example https://api.yolorouter.com.
A Yolo Router API Key generated from the console.
The model ID you want to use, for example deepseek-v4-pro. It must exactly match the model ID in the Yolo Router console and support tool calling and streaming.

Why anthropic type is recommended

Some models, especially reasoning models, require reasoning_content to be echoed back verbatim in the next turn. Copilot CLI's OpenAI integration does not support that mechanism, which can trigger 400 errors.

Using the Anthropic Messages API compatible endpoint avoids this. If you only use non-reasoning models, such as gemini-2.5-flash, you can set COPILOT_PROVIDER_TYPE to openai instead.

Setup

Check Copilot CLI

Run:

copilot --version

If it is not installed yet, complete the official GitHub Copilot CLI installation first.

Set provider environment variables

Copilot CLI BYOK mode typically uses environment variables for the provider type, URL, key, and model. Use the anthropic type first.

export COPILOT_PROVIDER_TYPE="anthropic"
export COPILOT_PROVIDER_BASE_URL="https://api.yolorouter.com"
export COPILOT_PROVIDER_API_KEY="sk-yolo-..."
export COPILOT_MODEL="deepseek-v4-pro"

$env:COPILOT_PROVIDER_TYPE = "anthropic"
$env:COPILOT_PROVIDER_BASE_URL = "https://api.yolorouter.com"
$env:COPILOT_PROVIDER_API_KEY = "sk-yolo-..."
$env:COPILOT_MODEL = "deepseek-v4-pro"

set COPILOT_PROVIDER_TYPE=anthropic
set COPILOT_PROVIDER_BASE_URL=https://api.yolorouter.com
set COPILOT_PROVIDER_API_KEY=sk-yolo-...
set COPILOT_MODEL=deepseek-v4-pro

Variable reference:

COPILOT_PROVIDER_TYPE: recommended value is anthropic
COPILOT_PROVIDER_BASE_URL: Yolo Router gateway endpoint. For anthropic, use the root endpoint
COPILOT_PROVIDER_API_KEY: your Yolo Router API Key
COPILOT_MODEL: the model ID to use

Optional: use OpenAI provider

If you only use non-reasoning models, you can use the OpenAI-compatible provider:

export COPILOT_PROVIDER_TYPE="openai"
export COPILOT_PROVIDER_BASE_URL="https://api.yolorouter.com/v1"
export COPILOT_PROVIDER_API_KEY="sk-yolo-..."
export COPILOT_MODEL="gemini-2.5-flash"

$env:COPILOT_PROVIDER_TYPE = "openai"
$env:COPILOT_PROVIDER_BASE_URL = "https://api.yolorouter.com/v1"
$env:COPILOT_PROVIDER_API_KEY = "sk-yolo-..."
$env:COPILOT_MODEL = "gemini-2.5-flash"

set COPILOT_PROVIDER_TYPE=openai
set COPILOT_PROVIDER_BASE_URL=https://api.yolorouter.com/v1
set COPILOT_PROVIDER_API_KEY=sk-yolo-...
set COPILOT_MODEL=gemini-2.5-flash

Optional: persist the variables

If you want the setup to stick, add these variables to your shell profile or system environment variables.

Do not commit real API keys to your repository.

Run Copilot CLI

Start Copilot CLI:

copilot

Or run a single command:

copilot "Introduce the model you are using in one sentence"

Verify the connection

If Copilot CLI returns a normal response, the integration is working. You can also verify that models are reachable:

curl https://api.yolorouter.com/v1/models \
  -H "Authorization: Bearer $COPILOT_PROVIDER_API_KEY"

Troubleshooting

401 or auth failure

Check that COPILOT_PROVIDER_API_KEY is your Yolo Router API Key and has no extra spaces.

Model not found

Check that COPILOT_MODEL exactly matches the model ID in the Yolo Router console, and that the model supports tool calling and streaming.

404 or incompatible endpoint

If COPILOT_PROVIDER_TYPE=anthropic, check that COPILOT_PROVIDER_BASE_URL is the root endpoint https://api.yolorouter.com. If COPILOT_PROVIDER_TYPE=openai, use https://api.yolorouter.com/v1.

Reasoning models return 400

Use COPILOT_PROVIDER_TYPE=anthropic first. The OpenAI provider may not handle the reasoning_content echo required by some reasoning models.

On this page