# Connect LobeChat

https://omnimodel.me/apps/lobe-chat

> Self-hostable web client with plugins, multi-session chat, and a mobile PWA.

## Use cases

Good for self-hosting a web chat UI for yourself or a team, with plugins, multiple sessions and a PWA for phones.

## Tips

Create a separate key for this client and limit its allowed models, so its usage is easy to track.

## If it cannot connect

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

## Setup

1. For a self-hosted deployment, pass the API host and key as environment variables.
2. On a running instance, fill them in under Settings → Language Model → OpenAI.
3. Add the model ID to the model list and start chatting.

## Settings

- API host：`https://omnimodel.me/v1`
- API key：`YOUR_API_KEY`
- Model：`MODEL_ID`

## FAQ

### Where should the key live when self-hosting?

In server-side environment variables, rather than having each visitor enter it in the browser, so the key is never exposed.

### Are plugin calls billed?

Plugins themselves are free, but each model call they trigger is billed at that model's price.