Continue · Laras Router Docs

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

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

Set up

Store your API key

Add your key to ~/.continue/.env. IDE extensions cannot read your shell’s environment variables.

.env

LARAS_API_KEY=lr_live_your_key_here

Add the model

Open ~/.continue/config.yaml (configs dropdown › gear icon next to Local Config) and add the model.

config.yaml

name: My Config
version: 0.0.1
schema: v1
models:
  - name: DeepSeek Flash (Laras Router)
    provider: openai
    model: deepseek-flash
    apiBase: https://api.larasrouter.com/v1
    apiKey: ${{ secrets.LARAS_API_KEY }}
    roles:
      - chat
      - edit
      - apply
    capabilities:
      - tool_use
    defaultCompletionOptions:
      contextLength: 1000000
      maxTokens: 32000

Select the model

Save. Continue reloads the config; pick DeepSeek Flash (Laras Router) in the model selector.

Good to know

Listing capabilities overrides Continue’s autodetection. Keep tool_use for Agent mode and leave out image_input.
If you already have a config.yaml, add only the entry under models.

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