GitHub Copilot CLI | AI Gateway - Orq.ai Documentation

GitHub Copilot CLI brings Copilot to the terminal. Its bring-your-own-key (BYOK) configuration points the CLI at any OpenAI-compatible, Azure OpenAI, or Anthropic endpoint, so every model in the AI Gateway catalog is reachable from a few environment variables, with unified tracing, cost tracking, and access controls through Orq.ai.

Always use the provider/model_id format for the COPILOT_MODEL value (e.g. anthropic/claude-sonnet-4-6, not claude-sonnet-4-6). Bare IDs return a 400 invalid model format error.

Prerequisites

GitHub Copilot CLI installed
Active Orq.ai account
Orq.ai API key

Model requirements

The selected model must support tool calling (function calling) and streaming. Copilot CLI returns an error for models missing either capability. For best results, choose a model with a context window of at least 128k tokens.

Setup

Configure the provider with environment variables before launching Copilot CLI. Export the API key first:

export ORQ_API_KEY="<orq-api-key>"

Replace <orq-api-key> with the key from Workspace Settings → API Keys.

OpenAI-compatible (recommended)
Anthropic

The OpenAI-compatible router reaches the full AI Gateway catalog. This is the default provider type, so COPILOT_PROVIDER_TYPE can be omitted.

export COPILOT_PROVIDER_BASE_URL=https://api.orq.ai/v3/router
export COPILOT_PROVIDER_API_KEY=$ORQ_API_KEY
export COPILOT_MODEL=openai/gpt-4o
copilot

Replace openai/gpt-4o with any identifier from the AI Gateway catalog (e.g. anthropic/claude-sonnet-4-6, google/gemini-2.0-flash-001).

Route Claude models through the Anthropic-compatible endpoint.

export COPILOT_PROVIDER_TYPE=anthropic
export COPILOT_PROVIDER_BASE_URL=https://api.orq.ai/v3/anthropic
export COPILOT_PROVIDER_API_KEY=$ORQ_API_KEY
export COPILOT_MODEL=anthropic/claude-sonnet-4-6
copilot

Replace anthropic/claude-sonnet-4-6 with any Claude identifier from the AI Gateway catalog.

Configuration Reference

Environment variable	Required	Description
`COPILOT_PROVIDER_BASE_URL`	Yes	Gateway endpoint. Use `https://api.orq.ai/v3/router` for the OpenAI-compatible router or `https://api.orq.ai/v3/anthropic` for the Anthropic endpoint.
`COPILOT_PROVIDER_TYPE`	No	Provider type: `openai` (default) or `anthropic`.
`COPILOT_PROVIDER_API_KEY`	Yes	Orq.ai API key.
`COPILOT_MODEL`	Yes	AI Gateway model identifier in `provider/model_id` format. Also settable with the `--model` flag.

Troubleshooting

Requests failing with 401 Unauthorized

Confirm COPILOT_PROVIDER_API_KEY is exported in the current shell and holds a valid Orq.ai API key (not a GitHub, OpenAI, or Anthropic key). Run echo $ORQ_API_KEY to verify.

400 invalid model format

Set COPILOT_MODEL to the provider/model_id format (e.g. anthropic/claude-sonnet-4-6). Bare model IDs are rejected by the router.

Model returns a capability error

The selected model must support tool calling and streaming. Choose a different model from Workspace Settings → Models.

No Traces appearing in Orq.ai

Confirm COPILOT_PROVIDER_BASE_URL points to https://api.orq.ai/v3/router or https://api.orq.ai/v3/anthropic and the API key is a valid Orq.ai key. Requests sent to the wrong endpoint will not appear in Orq.ai Traces.

Verification

Launch Copilot CLI with the provider variables set and send a prompt. The response appears in the session and the trace appears in Orq.ai Traces with the model identifier set in COPILOT_MODEL.

AI Providers

Coding Assistants

Frameworks

Automations

GitHub Copilot CLI | AI Gateway

Prerequisites

Model requirements

Setup

Configuration Reference

Troubleshooting

Verification

​Prerequisites

​Model requirements

​Setup

​Configuration Reference

​Troubleshooting

​Verification

Prerequisites

Model requirements

Setup

Configuration Reference

Troubleshooting

Verification