# Connect Kilo Code

https://modelportal.ai/docs/integrations/kilo-code

> Open-source coding agent for VS Code and JetBrains, connected as an OpenAI-compatible provider.

## Use cases

Good for letting an agent work through a task in VS Code or JetBrains, reading and editing files and running commands. One task makes many calls, so cost grows with task length.

## Tips

Coding prompts tend to be long, so pick a long-context model and set a daily spend limit on the key.

## If it cannot connect

1. Use `https://modelportal.ai/v1` as the API host, without `/chat/completions`.
2. The key starts with `mp-` (older keys: `key_`) and is not paused or revoked; `invalid_api_key` means the key cannot be used.
3. The model ID matches the [catalog](/catalog) and is in the key's allowed models.
4. `insufficient_balance` means the balance is too low; see [common errors](/docs/errors?lang=en) for more.

## Setup

1. Open the Kilo Code panel settings and pick OpenAI Compatible as the API provider.
2. Enter the API host below as the base URL, your key as the API key, and the model ID as the model.
3. Save, then send a task from the panel to confirm it works.

## Settings

- API host：`https://modelportal.ai/v1`
- API key：`YOUR_API_KEY`
- Model：`MODEL_ID`

## FAQ

### The model I want is not in the list. What now?

Type the model ID into the Model field. The fetched list only shows models the key may call, so check the key's allowed models if it is missing.