Goose · Laras Router Docs

This guide connects Goose to Laras Router. Every step was checked against the Goose documentation.

You need an API key (lr_live_…) and a topped-up balance. New to Laras Router? Start with the Quickstart.

Set up

Set your API key

Put your key in an environment variable.

Shell

export LARAS_API_KEY="lr_live_your_key_here"

Add the custom provider

Save this file as laras_router.json in ~/.config/goose/custom_providers/.

laras_router.json

{
  "name": "laras_router",
  "engine": "openai",
  "display_name": "Laras Router",
  "description": "OpenAI-compatible API for DeepSeek-V4.1-Flash",
  "api_key_env": "LARAS_API_KEY",
  "base_url": "https://api.larasrouter.com/v1/chat/completions",
  "models": [{ "name": "deepseek-flash", "context_limit": 1000000 }],
  "supports_streaming": true,
  "requires_auth": true
}

Or add it from the app

Or add it from the app: Settings › Models › Configure providers › Add Custom Provider, type OpenAI Compatible. In the CLI, run goose configure › Custom Providers.

Start a session

Start a session with Laras Router as the provider.

Shell

export GOOSE_PROVIDER=laras_router
export GOOSE_MODEL=deepseek-flash
goose session

Good to know

Goose’s own example sets base_url to the full chat completions endpoint, so this guide does too.
Goose ignores an API key written into config.yaml. Use the environment variable named in api_key_env.

Settings and screens can change between versions. If something looks different, check the Goose documentation. For other tools, see all integrations.