Install Kilo Code and Connect It to TeamoRouter

Kilo Code is an open-source AI coding agent for VS Code (it also has a CLI). It lets you add your own provider and choose which API protocol to use for it, so you can run Claude, GPT and other models in Kilo Code with one TeamoRouter API key.

Field names below were checked against Kilo Code's documentation on 2026-11-03 (Kilo Code 7.8). We tested every connection on this page with the Kilo CLI 7.8.18, which uses the same configuration as the VS Code extension.

1. Install Kilo Code

1. Open VS Code and press Command + Shift + X on macOS, or Ctrl + Shift + X on Windows and Linux, to open Extensions.
2. Search for Kilo Code.
3. Click the dropdown arrow next to Install and choose Install Pre-Release Version. Kilo publishes its current extension on the pre-release channel and recommends it for all users.
4. Click the Kilo Code icon in the side bar to open the chat panel.

VS Code-compatible editors without the Microsoft marketplace (VSCodium, Windsurf and others) can install it from Open VSX. On first launch Kilo Code asks you to sign in to a free Kilo account. You don't need Kilo credits to use TeamoRouter.

2. Get your TeamoRouter API Key

Open the TeamoRouter dashboard, create an account, create an API Key and copy it.

3. Add TeamoRouter as a custom provider

1. In the Kilo Code panel, click the gear icon to open Settings, then open the Providers tab.
2. Scroll to the bottom and click Custom provider.
3. Fill in the dialog:
   • Provider ID: teamorouter
   • Display name: TeamoRouter
   • Provider API: OpenAI Compatible
   • Base URL: https://api.teamorouter.cn/v1
   • API key: the key you created in the TeamoRouter dashboard
   • Models: Kilo Code reads the model list from TeamoRouter once the base URL and key are filled in. Pick the models you need, or click Fetch models to retry.
   • Headers: leave empty.
4. Click Submit.

Model IDs you can use:

Or edit kilo.json directly

The extension and the CLI both save providers in kilo.json (~/.config/kilo/kilo.json for all projects, or kilo.json in a project folder). You can paste the configuration instead of using the dialog. This is the configuration we tested:

{
  "provider": {
    "teamorouter": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "TeamoRouter",
      "options": { "baseURL": "https://api.teamorouter.cn/v1", "apiKey": "your TeamoRouter API Key" },
      "models": {
        "deepseek-v4-pro": { "name": "DeepSeek V4 Pro", "limit": { "context": 128000, "output": 8192 } },
        "glm-5.3": { "name": "GLM-5.3", "limit": { "context": 128000, "output": 8192 } }
      }
    }
  },
  "model": "teamorouter/deepseek-v4-pro"
}

limit tells Kilo Code how much context it can use and how long a reply can be, so it knows when to compact a long conversation.

4. Start using Kilo Code

1. In the Kilo Code chat panel, open the model picker and choose a TeamoRouter model.
2. Describe the task in plain language, for example "add error handling to src/api.ts", and press Enter.
3. Kilo Code reads files, edits code and asks before running shell commands. To change what needs approval, open Settings → Auto-Approve.

Common errors

• 402 "wallet balance is insufficient": your TeamoRouter balance is $0. Top up and retry.
• 401 or "Invalid API Key": check that the key was pasted in full, without spaces, and that it hasn't been deleted in the dashboard.
• 404 or "Model Not Found": check the model ID spelling, and that the model is added to the right provider (Claude under Anthropic Messages, GPT under OpenAI Responses).
• No model list after filling in the base URL: check that the base URL ends in /v1, then click Fetch models. You can always type model IDs by hand.
• Connection errors: check that https://api.teamorouter.cn opens from your network and that VS Code's proxy settings are correct.