Laras RouterDocs
Console

Integrations

Kilo Code

Open-source coding agent for VS Code, JetBrains, and the CLI. Step-by-step setup with Laras Router.

On this page

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

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

Set up

Open the custom provider dialog

Open Settings (gear icon), go to the Providers tab, scroll to the bottom, and click Custom provider.

Fill in the dialog

FieldValue
Provider IDlarasrouter
Provider APIOpenAI Compatible
Base URLhttps://api.larasrouter.com/v1
API keylr_live_…

Pick the model

Pick deepseek-flash from the fetched models, or type it, then click Submit.

Set token limits

To set token limits, define the provider in your global Kilo config (kilo.jsonc). This block also works on its own, without the dialog.

kilo.jsonc
{
  "$schema": "https://app.kilo.ai/config.json",
  "model": "larasrouter/deepseek-flash",
  "provider": {
    "larasrouter": {
      "npm": "@ai-sdk/openai-compatible",
      "options": {
        "baseURL": "https://api.larasrouter.com/v1",
        "apiKey": "{env:LARAS_API_KEY}"
      },
      "models": {
        "deepseek-flash": {
          "name": "DeepSeek Flash (Laras Router)",
          "tool_call": true,
          "limit": { "context": 1000000, "output": 32000 }
        }
      }
    }
  }
}

Good to know

  • Without limit.context and limit.output, Kilo turns off automatic context compaction.
  • The environment reference in apiKey is resolved only in your global config, not in a project-level config file.

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